From 1e91a6ab0ce0c33429f585457b0b6e54e864e4e2 Mon Sep 17 00:00:00 2001
From: Krzysztof Smusz <ksmusz@habana.ai>
Date: Wed, 8 Oct 2025 16:17:59 +0200
Subject: [PATCH] [Docs] README update - bucketing, warmup, defragmenter and
 sampler warmup (#353)

Porting https://github.com/vllm-project/vllm-gaudi/pull/231 and
https://github.com/vllm-project/vllm-gaudi/pull/278

---------

Signed-off-by: Krzysztof Smusz <ksmusz@habana.ai>
Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com>
Co-authored-by: Yaser Afshar <yaser.afshar@intel.com>
---
 .../graphs/exponential_bucketing_example.png  | Bin 0 -> 99919 bytes
 .../graphs/unified_bucketing_example.png      | Bin 0 -> 24883 bytes
 docs/configuration/env_vars.md                |  58 ++---
 docs/features/bucketing_mechanism.md          | 186 +++++++--------
 docs/features/warmup.md                       | 225 ++++++++++++++++++
 5 files changed, 331 insertions(+), 138 deletions(-)
 create mode 100755 docs/assets/graphs/exponential_bucketing_example.png
 create mode 100755 docs/assets/graphs/unified_bucketing_example.png
 create mode 100644 docs/features/warmup.md

diff --git a/docs/assets/graphs/exponential_bucketing_example.png b/docs/assets/graphs/exponential_bucketing_example.png
new file mode 100755
index 0000000000000000000000000000000000000000..02dd0c2c72668f510703ff8a2d9cb9d7006cc78b
GIT binary patch
literal 99919
zcmd>l_g7O<)2;<2s6YZrks4Y+nh1s#Y9fe~fGDUm0TB=)B2oeb6a$0~Qk7m56_MVg
zN$5f8B?JL!(h}({5V*nj{nq{dfcx9sYq7GDlM~LK*)w~dXPzBqY^2M4?((@4Cr&Wy
z-9+3zapH{Di4&(v0cYu->?;errN5l?xUH*w;%hhWBK-}6qn0848znL44{XoS-!nbC
zdEev22@cWYzms~mMOIIoIC`&#(7NMeO&<3yb2WcEzb7p#G8@Y&C<_)zi)}h*Q+-AJ
z<I<DKfq|7g>t@7{MtP7LNWM?*`oPecixW3bVc7+llpn?FT5C_p@08W($9pP`EM@p1
zKQv6Tt6oF?Szo@dx<jQn7S>M+R9(G+_@D7beBTWdh5MgzPmfJYSm?jQ?!6wv|Gktb
z22%K+D?(=P5WW9%5AL4x|6eyfZq#^cyS)Ez{u<45(_ONSFg4odz(71wU76B&bU45>
zx!bn5+RU2CqV)S^VYblr1y{<bV`e()r=M5b1dGZ-Rw4M<mR{-q%y<d=Z1snHk3o;x
z&aZ<2bLHLNuk(eYm4y1hkF9+d-!HNT?l!*|vGO+D`Q_2`;`b{?qx{SO>WemRM}ts)
z--Tggza&l@e|HWm8=tA*`%Y@);^xP-zb^G{C7N5monar$2t3>?%>2q~JCgl0l=pdu
z%zVXtV@J>H+ZEX=byLi4DKWI|F*lcwKO;Vc1vol-|8j5t<?V)9w#YB*QTJ1!F635)
zgwv->^!(>&Ghon==ZKZ!__Jo_0aK(}MVGdC-;}VxUdIc9M7}Fc#JUIgp4i_LuwAy;
zu>Uu;6@Ol~?isZOqF!Q)AEfO}YS5-R8&iVKgAQsZX@BqgsI7MhSA`Jeo-*K$QYQ}{
za|~8`j_UJ@_^v}?Yc1^g14gPS-!6-Z(uAB;mC1m8ibtT@!D_Qd`ovna_m5DOr(S6V
zo1=Hlt*+Cys~XdnJ^t3r8)Y8MLLB?l=vKD#R{p(GRlJ3vFZwS4Y|nC0V*qulC{S%2
zvol_nwfXxs>)*{Lp6J!p4~q59F5x9lR%87avO_<lKK<tWZJXrc=pH}v+gC_920wXp
zxHEvN8i%T_em`C2xS$6~S4j69zF!tPrK9<Me@-?~dE93vs?1#2eVu3@ldh)i&9U~k
zerKX40NcCeI-bqPyY9c)r#qhtfB{1pm#(3)R}`DfF=5%H-@7w0nQpFJBl|&9dS5J`
zb{7jzeAeaj9L%hc&VC^9%XhhOqo%9o2O0-)n$M+s!$*4aVIKYU5;pdGZuss)^cT@C
zERg?c-J;ivH2?KZp2{D}P)39?{!dTf;l^U-(Ozc0z`<|E4Yt`W!3;jSmj~WDW+<P}
zoKN%UCX3d^jfr015&O=fI&n7-Ji&EdFt4z}`&2e{?U`ruODAf4=D{NlGQS_W#C%5P
zbZ$2R#*#_VPta0jd)&B@zG5&*9j<zeybG=7u&#jxBGtAA@9#cxnOF(mtyk=AV{qIO
zBKbBJ@b7>?pb~FqWq$;s<s80%vQIbZX08Cui6ah|xq|NTdu(FsSl0x|N;);hBWD&5
z$?&n`$oX2gPFQ^B^7XBe(iC>F=!jaB9}}>dYXLtv0{R2Yl3F*UMD9d|4!I^I&#naQ
z)ZR}MO<Oxrv3Y2<;yWT`_E#!5+T7B;Rmzo3Q)9a%#*owMMc*(mP&?hHdcuF>=Z4mx
z5H7=e0`1|W&0Ey!@UmT$^_hL<$HH#vW@(qqn4==Jd?np`Liv1`25k{J@9~BLC5~0X
zVn*M+Uv-VClCymwFFg<OW&44?%#LPStV@gvJC2vMS*FYM=}rSb0n89G{C64p`1r&&
zn6l&q0c^dvIJba}R7cs~t)y#iUOrzxH$U*3ANOrK!)cIg`Mh(p+q=PoqZHQ=16%JP
zOY28Q)Q1K<XAemqO}(3m;y9U51xXeB^Q~`I?9{7t!Ed>tZuG3)(jbaFPRy4c^|2>F
zhQ=ngD;MDc)ZT+<3F%gNGFuZi)A22aJxm)`8m;^%Wz^5bMC4ME<3(1L32#`g&r^^!
zQ!%X8|F56l!F|##C($ozd$aKyj#x4D>a{yMwg520CraXdn$Nqa7j?;vJz?Gi34u9o
z=RJREh6_2#bA`rF_t>c%14%NtYyB4=>+b7(PNu4s57_M`=vqh2S*%D*kV1Ebn78^^
z*AtU^boRBrQNxWt`NK8)GhyUGlLz)IH;eEMN3@kp)FWI*9iUx3lB5aS+qFJy`LspV
z2`r+RZ#h7wuf(X7342=isG@~oTBE->8+ThZHoWf2P(HfN?nJry+i&L#Lh5wTCefkm
z7yJe|Vl)nSd65=GZM*tPE=<pMMZNDB($j*~=dyigHBUwrDc<mcU*F2KzDr>}`3qzr
zM;do}k*QJX57soJVn#p1bz+yz8l9&>y{*b^O2Q=<mJ$P#@(UG4NB1Xo8kCXl#7t?`
zyec1=xajh&T3`DF=gpHuT=P6(v3DE)L~-#}yAN?9etpQ$673>@l8rL4Sv`CNS>LHY
zypvqjNTU=*Cm$~te~};^dWsVC(`L97lcibyn=+d%hXo+dXdE4E90haT5D~mUv{_HG
zc>1=LmDP;88ma-4Wf#o?YQiXE%?|)NNzH6}YA0H6-)H>;h5@B;e6a7hb9I}5^UGn8
zgqx3Mg0J_3A5kMyvcF<k_kbm2XFYN7@Rv(kgE8L8@2C1NS=Ta2qL5#bwpX`>NEcYO
zC$QM8is~Qz(CJT&m0)$YH~vNk;PN$}vb(QfVG0bX6>}T0Tz(R_bTKgL^xYhk?zag^
z)wsB<JlbZ>4hb3fnK3RBCy#|@N=8&(K9vHSpqr~(D9hLBbwN>s?_bBcc#SFDv{%p8
zArG059fL)L8&d1Ej!@=e(3W*i!3($WtkVuVi1ng?t%8k$Yik#1gZb*$lP_Vf=kTk#
z7hkuZmOY%7{o^Z@(?Gh%U7lV!A4QMOK;O#3ijGW_ll@HevxT!*81Z$$kOnE5Qo?C(
zZvq(S2HmLvU`Y4)HWPQ-Le;B~pZybbl@=rA9;S)su99EYmj+s#<tn2Na}7^+!>K)|
zzm~uCRhrd6teQY7sZFc_&;Hs#XDqXsBNfNkc8#c9sJx7AR3g8uHSHm<QahI{!)@Ao
zfr{7HE*Ev{ty`(KMe5-{>@L+h)xzr_S(oKqyQ=L{sqnq18So(8-7SA|^P6>#-Htsw
zfx(_Qb<+0Lt~T)j(%=_A0vL1}j6I?KOm1|Z76SeSyo1Zave})~3MOj)7A%#BfMCw+
zW{G40b?6qd`H2tq#|lpaWkW{9*Iqr;_FoMUwAo6SINpfwyP7zz<so}l)lT+Lc?5-*
zCpe>2f!6_%(cm^VO&pj(PP<_Jrbj|w!8iRF!BVq3CM*=4UNG@yM8`Eke#4=F!yR>J
z(Cy(Q<QRKDAY<vPKzXV!dtjpOXhNUgBAZKs-n$kvyXck8Gp}?<F(0JxQLxbqO_9}b
zpEsGSBppnGDJIN;>X2m(-?ckWGOU@?I!Fk;Mm12PMD{@<#Ph$govdJNiow-L3dKJs
z=<XL^M*{^C{nQfNX1lD*87&;Unwyb)p~fH7umc>}Z>^PFAB*Igwi92EOTPa0<;EoF
zu$<2wso^Y4nP06x2vMUiIE&5icN{BKrmw_P*{<kv(f<0`#{fQeGgt<de=EJ@>3BqF
z4hXhCO%Sx1o)+&}(04B{S<}Fpxf6c?*;|8)+ftrQu(Kzc9NjW)XxkP~JrtPdT`UIA
z=qcHj7hD6?hIKcXBOO-jOFl=sTJe)wxi0To)^g_~ETDpQ(18ZRle#B<G+Aavu3r~=
zA9t4sThgl!SEK(HQ1~vFOIl3YVw(uMzrnARP6qX^-V?bAocY>#NL{QPdzv}}M8j};
z0Sq8JhF7luh_u*B1*VIF<sdekU<=S0C<x%}hkWVq0652rxd%Ie-E6}YTiEz*`h}-K
zVj(_S7w-g4@d`_epYOZzuH_<LU`oe=%@**jKS}H)mY8Rl(7c0?p5qb1DFA#JwCv(5
zO<DQ<>(1#uKuFnQWHs=Sn+a|p{CTjUDgnNiFe({iyr{fc;k#$VqpbmG<`87{?1iom
z%N&}^<U|KNE9S<>Hwe7|Z4BO5+j@Jz5v+mIf?c+&eh$zT9E9}WQO9I85`)S|-yCl3
zEwJyA4MBIt#Lrs{cT??WCF`zfb5T^DUMNOZHnBfb10_)OO#N<}s-cBt!n4*njJQtm
zruISbrJN2)j3q-U_VHvbfqff+5@A;)!%zdE$S^FpE$UhcE3JFLtOgq&y)NbZhz+|X
zT=F6fNjz11m*19nai8y$u*KHf^-uY}T9}ZR;?j&fmts&0@^d2ZsFtc1K!c8C`2I5E
z-epEoF!Pe+)R14=>++&)wq(BIW?OxGsS-uSJ9!1Xgf_3CWIlcKd!EK_GszeyPFmLZ
zaecY*25`L12a%aT7a1f@1u<#Fy>t>qIRjOA)#pL&rjcCQT!J5fPXPk`_auX&%fl`+
zWJkz0_-ua`vh-PvJ^(0T6T!V8M}W$@7y_t;yRm*LS`!>qQY2au0?7G`Wwm2<zi}TB
z2`6*sr21X}-Qf$fSTg!jBJ%5+?brz?-L<eQOG2olE?rFMZmu*=w|U_19!{MANe7FB
zF?z=Go7}Dy&n%5v-wVKH!GDOn0Zo=?I*gGl_u*Eppv;@ORzv1IZ13il11Ul_m`|?u
zNBUc4ET}ALFCB&<G1pfR{<W^1BncM{OhE%7Cp{1TU5L=Bi{?;U-2MRgtQI_EjrI3@
z=#9{&6iA13RD*=XH&0sd<MXzk2{$F3rX~qO1ZXWG!C<nBEQ-)jXC8wJ%W~DKE$u2n
zq*E@-uD?;=cNpj&j%m3)qcthf#l@JbA?<9=Px)ftN<i!QElSSu*~3$pkaMvi(kz>y
zFWPq7zJ_PDMFxNo=e!r+CwJT1r6KbQq$Bkx%`7^ja!>C$`PpB13gBX;-rd&=Rdcd;
zmdO2frD}Ueo$f8DV&MpVv&B*Do+BTE@xTTEf6^&-3BXD0MLei4_~kVQPJA=4d{~G#
z8_V@bKqH93y+g}YkN^aSB)=7{U<qQwaUg_qL_rAdImwLYY+`vg%2zdBg$ioJz&9mS
z`*c*6wer9upcDE1Er5xP9ZOu_e7~?D=uu+uYI#aVDM6R+jOQO^<&ZTANagqN+5WR6
zv%SH)V^>yuoj57^6mp_CvolCBbY9dpq&lCL_bD?kpE4AsH0Oz{q!Nw@H(Z5)Akk1j
zt$d_{4iB@t+5OR2SfVM9xIo{n^a_EKZ(=KC1s}0B0bavSGPx2yNLLg-cN3xNWwe3T
z+<h*vpBuei>)|AiDKOjunV2+|stFxN_YPNz|Nhh^wwdg|$$m27cVXkG+^aZ%;fcg5
zx!=8F_FtA9p_C7y(91z9Tm|F_^mp&JaA;yajL4B{%v!`SFn)=x)s%luxVlWu-nIN~
ziKD+ng!}uVu<X}hUdbvsA!&+GhC;%)&+0@?l(=Nt3nHajf(`^gsj2?~!0il^vlFFe
z$olQ#w<tdTIym53jm^bP00-hb7yz>Isa#8$xvR5#!u=r*I>xR5aDZJ8gW=h1&LWgg
zoh=F06yn35>3a^~w5B0n3O)ph16-!1!i(Su8#T{C@*%%(Y~H)5&0)en_eQJ_PQ~i_
zh~9Q)M1x-B!8gG;Hbgm}i9UCjDZVRHP?*xX4hk$!h2L)#J*Sf3sn){nDRfoF*EqVA
zbt<~}qic)Q+}KT=C5e1mVD~#Ys@=MLJV)B6=zxT%so?Z`?)qYHwyDg$a6FK<Y?Ilr
zQ@`@l9xGkOKFx2EE-gcv&v@*U#3nx)p$@?_msq9-rI)DDo{(3GG_s?Hs-I+OI{2~M
zc~3|w$t5AVdNU7xQ?YWZ_2u{W!?tQMOLjfZ_LH0MduBCAdRBw4e^R6et@v=wGrT7}
zd2<?{`?~pqUUQ!D_R6n=R}USL(1lkE0-h=?2zW`>ZT{PMl3xGU+#=}!!hQsRT2X{S
zrM6enK_a8po2zgiVX@Wk<^IXQTdTq>s}myHK)w+1Dt-n|>qjk#H%!h<g93a)K|BCf
z2AKAOI6DZQfdH(9akOaSTS@+*AGPx0GLBlUmYFL%fT@FTfu)s^EteK|f9*E`;R!LC
z`X@vyon)tm;0+8#?!eR#&P-qZ*Rgq#W9yNFVg#L=cP3k}b?B<B^WOAu16*GpENT(2
z%!<r)T_2f%nKC`cR`!`=HQIHO-EmpVqqJkV@29k^x~+Hf@oPePVE~c+0w{s6*fEHA
zY#_R=&u-GU=MV8(+}=j!(ZOb1jQ=D4c=)bG#JYa6BjjWP@H0hM1Ed|AqG5cN_UK1g
z&xHTun*AJ#3Lt~*ORTQrXkVIm$;Um0HJ<S-ky)+xvVbXmsknZm+H#k*-FML`Lawqr
z?3gP1UpD=&fNi1hwQy<oJ<tm0(!3}9Y&}-EAE*Jysw3bm8fKJrBP}oT48wil2KOWt
zH2v{ERwpiJB*7zvra#6!r)I2X`(q<2Zf2g)zA8dsLjLBq&RYf_fw5!*D~`UW41)L%
z=4(tYGue_Fd!5Kb^6ytpiwCR5rslL_={3W5E;rI|BskZ5jXTYLve69`p9F@NSG=D<
zD2;Ie2M2q+;V(%}vsH!eSIHTLpwDQfF@fiPc6%9`w5|POH<V7fx2YD-*Bd-#wzh!4
zw&@(m>VwEjcJFB(L%ISt#}3PrqS$>$qIu?w#ZavwQJ0W*-og%Q`=Vdj+=D%(qKE2V
zLFM>~6WeVq+pG2V#LxDHhM(D0-5;h2cqMopmp&|&fp-PyA+OK+swKcj;BYlGzmYc9
zn0~m`csMc!YdDyb?eQ~+rd{Tg>w}1_S3cj(NV)k4?EioYgDLJXw44!o?H>ke2|Jk`
z{BZd$4WOk;Fn<7lA^P|W<T;rsPZ`Fah_q(VV6_4Vef3^#m<v|I_Ec<HK?l=+Z#N4Z
zKHDxo2o2l~-P>I<Z&bnT)J~?;2?T+kIq#@{fBLeqppCye?gwrt2Fha&W&|?T?u}Mb
z*kFT8O9e$4YM-{|bg~+m{*doCZXDj}1`9QEluLu-zyAg6wiqaK_ynrpvRyWed!|~Y
z)U^hNIY~i)Zy7W;E$MVlFXCUF$igT*x~ZlsTLaL#L|;ymLp}VB<R)|rx?NfBe=i}1
z%$ti-q5}KG_Q%<d7NMB4(s82ExGJ2U1~_h^LqC50Qps!FbYS~~n9``A)myQvpVM+-
zsb@|ti%&WUbQQEuWV*2G$E#s)jBWG^x7=T^t7L`BBV;`*$<&d?!!0s+cShjIhoigg
zn(IV!bHVM};!pKVtan*^?)Pj*G3$3xN*1T!csA{@8iprh?5@~{l0T=IKC2PK<tp<@
zkMod&k7C+B@|=y{YLo1X5!8pB;hB~GmNBkE{7Tavl^XLTLSLuUN;#dM?^xT!Ov8Fm
z77*ZaSVvtlx`91DENi-&d)Bh1ilRiHvscW2Qc);T0jq7TU-zTZb2sAz+}P61PJ`i6
zS7?|>{~zVKKx@;6F+jGH#Ra#rpMN}|byMGSV<3Ma>T6-6?B4Y=Hnf7k>$LGn+W4m1
z;m^jmN1WT_vYw26yZeF%4^)gc59lPy#uA(J7DQ(BN>h9-ot_z!Zhh?Lr?155cQ<3p
zb<&K9KWK{%;=O$*_a_v0xCdXIc&w*vF$svQhywQKeQH4baJO&1CwgaD(r%JeGIKN)
zbpbWAxK`IPLZT=US5`mL`jZ<J&Uda`aHOqlHmXBhpD5xLK}Q71qBOPSqU!1JJwdF>
zPR*yL{dAviAPgrx2K+MqO`a$L5Ki*kSP%xh-HmZrsGB7-FWKrfzjhUlgSdx$FRR}4
zm%oaIoD%4Bma2u65+6U*W$aNobsqVxUIN%)w)qO9KvC;7H5zBi{Kt06>UZ@rMZ(^o
z>z}Z0jY*w)^SoGcuAi}vcQoZt82dB<@)63V|9g}v=BHYP;rEXNz-%-pPdmccIy%XV
zM<YU8d@7wL!P@5iG{+}))E8Ro<D`fZpLEbt1!)hN*c4l>;zWH(W#)eOg^H|6Z(@7w
zQ@$Hm#?Wi2pU6fceOe`O0Jsho20@&hc*JmtRj+h2@v7#D3T%S$H%Re|<YK*Y@I`6g
ze3mBsJc1VwDtTfd-dHZBEG7ezkoKMsk2CQDnkJam;2sdF?_)_(FL%{o7b*%E9n$Ii
z^~TiVT#MYqD}=E2m`(-FB{=K-5fLC4Nhy}YREcx^m6&Z$8WrA$b^{2c0WP0HC&!#T
zHO(-uw!(6>%7Z1hUos@oRMD4L;!7n?pCp!;WEAhe1HUM<!RJGLd;BKk&+k~L&jT`H
zb{ea%Sbu~bp<l5y9rrt+8ev)griNqSPqMpTOj`qlVGH$TY939Tj^)yX{T3yhk0n;C
zH96tl^jq<CAGL63M=1nz=ZL;K3uoB6!e`5e|9Zn$9U}GVv`J_OoQniF`8rnAN=RJJ
z@R0Dh_JsC=uM^T1AjcvkKZO#8u_~UUW&L8X!N4~kal-5>PRn&?V{`;oTxnR&;bk3p
zLfhkE(mZgAliOj;Y|sBIs!~*t1@yw_@LZ4C=t->?Y6)FG&z|~#(GQ+XdieQATg;*0
z%@`C%<QYW;pA|MGUH_>>I9KTUuJz>80v5uIp98;F1BBsMeQ?R2y7y{ii!S!zVdm{{
z+QGbr24*e7Chc&k=<o@3GhahZ<6(3PtXp7t^tgHSUnpxA3cN1u4SCM?7<S_=K&7sW
z{9e~->deiXf(dLom1p|2?gcT$Ie?D3UU$gz=fUZeYyd6^kz4Y}-{<rF&sVxhoBl`x
zzC8&}H-x@lA{q(5XkvZpw|LpMu$jcXFR;_+5BfRvm?GN?o5k<@-|ziPD5wu}qT+UI
zygpq88~On;-ewDlABz3p_e|BbWl~LaK8D&`h$pKQLfBZ`{qia}v>Vo1_?g<sq3Nz}
zE=;FB&?Jsxj%JQh4)+6(RyJ;<gk~#y8C{=CML@7Y^SSjFQKL7;1_ND_=VffYaU07+
zJv#(?ypwm1nd%p`t&tu|mDER0@O>Vq4fnr7r<Qp#vZJRuzetB>9dB=4a3QB-^Keeb
zt7%GmLVf^Qn6mf<wXw6H8<r&kzkwS*k!){<*$sc!P9APo4O}0wl+Ra&O<*B^fDTea
zq6WasJZa?gX&gIjj6B{^VCuVGzx=iKvD_1YOTVydhyLDN?hpH(_9&p1)~<+<gyIyU
z0c-0ZA|`#VrJ+nwSdL&i6e@V5nshVH!DpX&g#>0m^~|{WNuA3iYpUehXi05o#Hf0T
ze%=psL>MoEhKSQTS{))eE`hYGuFkjwU4HcJF9_V2m(}xj0EX}<bah@HwfNyA&rbT$
z2;7$(`?1G5&+rRp@=b|})vH^dBM~~Nj)?eOSq-QnzA)53VKm@NT~*Zm%uo5{L<p;q
z^8$QkEX7oJ&bC(g$IJAOWmJ$>Me0>-OKX5bd;cBe3I=jPb@GCCql${GlYm#2VE}a?
z(`fSGk7E8Y?N`KavriiNaNLX<m`k=%*EnIY&;GJ%<nM7zqLuzvUx(qW5W9QSZymCb
zF-RJ4xw*{6f^jX>N$Z5ronQ1uRoSW?9;vzcjK}ZQt@w{qf=*EoqX_*29{p@gds&>O
z5GLnUDX@YU3v}S0W|a&6o<)XTnRB#JKNVMpQ?zasB>?Rb?tfF?$?$23zvBy=D-Wdt
zy(ad$y}OGDVz4~;4>0W2JSVI9Ud+|g+DvPDB!Z!43F9AN$<?KgY5;RWp1QiZ&7QMG
zPS&=k6Hkr|s|7M+*lp<?3duDb&7eG&oC&p#*fxXWW!(*wJf~Wrb<T-!r7b~3!3^+5
zVvjc-P}D$rqm=gp|8b@G<VS%LOMxfswr$=FCaA2h49=mx9dM;e?F3<inDs@C?Q3<f
zn4CyvK9r?Ip9-}Wb{iGKhsLn&X=zez^A739>vTXifpNQg8y89^)s;3g*WRa*9B~1p
zCZ%38>K&_!w>`naA8*CX&ufmkxXcZi;bXRK4lJ!UC7R;W!}D1qSyB0x_$#%g+?>>x
zg+A-`)Mfk7(vPQWFVnIDSSV%0m41*;t!$F64Uc&@bJc|JGOvy@Y;SV+eqCo4hWN4p
z*zL`N-kmIf4e62a&bpcU13tapt8bzGXI(i-S`EzsX9->Wqc!Srm8FZP@f(2LrzCq~
zG4x`F;m6eGeRu1v`tm`2zb9jSqxSS*BU}83R*Ss>N&q;P8eJ76tQw5<X)y?<fe^S0
zbr~%X?NRgCmzuo1LA}7*7Ga}~%<pm;*VM3Y!9mpMSY1~hq0>lteEvbO=~MT)>lv<w
zA5Ugue+%+W2P=dr3$soKZvr0F{TkF#nm}#KVfM!+)$^s(n=&}G5{LXwMU*;o=0tDm
z?hPitM2K&suU2vz&x4;t)$No6uArP$JwdwMOo=R+1*c!V=`9lm-PwO~ehoVmC>(uw
z*PNCeSRVHL!qhV#cG>VO1bmHe>1$w~bl^f`+)m{~@9D1i>eh+!4d0oU#$Pc}+<Tv5
z8g?G?<|xbEsjuJ}6Lu=|wvV{EKr2>W)YG?F*1+elk{($IEda$UvJdzOpJ8pK+on2;
zWscB}47BGGBjpt(EQ{n7X<qpuN|z5;(-yBH8M-oL&V~9);p@L{MC2^Fhpro`oW5Mp
z4I;?98qo48#CcT*V+tjTc1b$i3A&kk0dBY~f<%V;ZC#33^iiVAS}*^SR_|T)#jqk*
zdorEMZuui}tQD|3Blk)`=;IC+R3x?|Mp0EU_=9c$oy5IB2%lmBQ|ebRD^f3+PlkWj
zy10HzQ1AjU1ILO`?829b2eGN*^!rU^&@4?6O4`87`<&~~*`u!t$DEw91?XC1bUWfX
zUTLdiMS_@K1$m1$;f%G37RwbC`8RsOUoJFt9w8FIS>l=ITX#9MW2n*Z3DeSl3S|6l
zGbC1Ci)vsG#dPSt#l0kj^?s1u`niqhHL+O?Q>EH%o*n}teW{wMouN?@BM(4#eDSRA
zcYR4uH|L3P4?lXN<o$cz=59__2at8flHm*cTq^d&H=;nUD?tSTb%+mv-hpfl{9*8h
z=8XfQm&3AppE78!)!i~jx0o_k=J36jX5Rkd-@nti{?Jco0U#{vu+zBO%u-R(^FAL{
z&pEk!d9KfFX$pwsbr`nhwDy@Qn4a!9-#pCQFUK+qDFiFsdd=*!wKs85YZSYZwu~$7
zmcahnX*{Bpo#;NxOCq7;>X$-I7`F!6v6HR7{~RIoA4hNtB{0>5se?jy%gY#>UVS_v
zbY=}NlG*W$-YM_Ki-GAov?^AV3PR&J;njlL%%Fsh7i+DYJ=QTT961v{z^;zW-@gST
zrF;{a0j%r7{WwYHX{-WH$$BXD1mfzOBu4yhwDC2D!jrZRI8A~Ke>U=)OEZfW4w6fi
zkdF(i*8YAOm+#+ijGVhTk!^#87_*s`2ip&NqcX6$V2)0GUW|6}{VpKO`)pUd@7<8r
z03M8}#-h*A)1@Lx2tX?BfWAT}OIRHsu~FrD+5StCzZO^*nF?&e^a(Tj;Ob9h_9j1g
zFw4d|lg_f}^97a-WBdgDFY8yB9eP>DR0hrKiAsKJgpeT@RG@h6w=1vBF;~94AV!yH
zJH>~1^{V6bGS#TA(n}4MG~RwQo5h&&P&L+6o;g!<%v|7Nv8{WH&DhS8=-&m#J)gvw
z?b{i)0<Pxw{fm%%{ezDQ*)YQUqE?GB>sVHv!N=CRnc=y}iu7xX@}Z@B)~S2eyj+28
zlB$k@uG*=emy0uRVB<5ceLB5w`grRUcz}bqk&b*FwGz@B=<$$x!wT)95$c{FNhqWN
zN?5jTpGD;y9Pf3?qi$U{5US|oP7xQJK|XsYz#J@fp-Jm9gT{rph?G*ZS-}PX#6~Fc
zD#Jy0pH2`rtdGR@zz#oo@N@Eqb<DI9?y`)2p5)hSIQ@r$O0E%L+)a4Hgk+c9gZ|hI
z`}$Hi%mIQ8*MA8fmI{A&dh&d`s@Xo9b;1ysRskrxhdk0hy(D&M$30T8vI>iD1Lh~7
z)cSO3iH5bTCHhvgCDx64ew(RB!Es-e)=kVNOM<~-dhyayNq3dD2A~m`@IF@fi}>#1
zp15QM&MrmQTfX>x8x>))ucr;dXU<G<kqkiV+kDHFt+N?>e0ouG%x*8su!YZatLD)7
z>=FCqaG9i3Wwrw6z}@J$ZKYw-hIo8sVz-klt(wb7UHHp3_k37i5rN4zFP9uK<8qax
zkEwKsIgS3@<5U(mUu?1$F8gCa-aW7O-iNqom0OHU^Jx{<Hc{*P1AdebtaY~{M!tmd
zqq*3$wg)l~H(&JK;QWbS@kM>$b6s2g*H3-}IU6|5o|$sPbcIg(*m?Y8{6Rb{Zl5ZI
z7A6lVlid?^n6Q5i|B+^<#x#*pP={#Td@uXcg4q?>TXD3kaX6#Vz2C@Yq;a?w(_<0r
zb6hK?#-nQQ9v9PepI(Kjg(^zM+_BEsR;gg_@NE!+X;FSY(YT#4I=ykbBJo7b7#(O4
z>;AQf=_z)ZO`p=Po}V<ZaiL-`z5QFLL=<C+GAxVmI>mn;d~2Jn;>(0}*|v#j$iE)r
z`?=NQRWw!3t1bkUoH^J$CIY_AM63R(*OhjGxU#S!O~w4k>!-E`eW%5xGLOD9s$bPu
z?c&e*S7`l52*wPdXA`Jo(sy>KTg_b3VjGWJnwq9s#}gQqa2sv<VJ-Gb=Emi%JRfAs
zP97SW&IvsRP_q94P*bZ}-zb%1e?oD;=;EQW&>@FpE9iLu-Dq#X&q$1e&v#T;>O44Z
zFlDRn)HhFLXJ`k`#_4t=UukR&nmWdq1TB#AH~-}{;eSFjXj-yO0NR(Hz#f}4NybR%
z>-xpsM7VOK<@*Gy=%O+7#=zsA9yzP&lPdf<v%Fq(K~J~-Jcj?+?wSAKH{G7}anJc1
ziktr7e}<&tzw-URPEq*x=>NG=(*Ix6A75*^6@T3F`S19DeiM18UD_2xUm`^w*66uJ
z`c$mnvGTs?)f#xk$gf1Hnhe-DqW|v_H#y9PDs4JKw)<#5-HY6K^k>!0MBTwuXM{D^
z-%6kUoDKg}d)9Nu>N9?y->(*@LIpA)lKE-R37f}iCzaV)cxA}39Wb4AmDW`?y7p{i
z<$5$-FtLbbH%bXCP<-9Hd$7@i+?fnKvTQreqV!N^%)Onj%9%iCGKy@DE3fIh0Z03j
z$YW_te(m3HE*W3yC@8noZn|s%eX5MSdJ|19Y5$|5P8X659E;`BeWt?(=uf1^=ev+?
z0Y~vn(dBH5%3o*bdZf1yUv<S#GuI{4=?XYIl|L!=A97>VeSMpGCW`4iM>SmvC-As$
zv|!WlY3o#9E^D4H&Wkr-cgn)P?Q@g50R0CSGyML(r91>M{!_3uv#7edd_)N!vLj(~
zV}E91G}Ol+7SrQDlZX&My0S*tQEvKUZZzquW0KEwSr;AtJTe+HWSg*>#&u1+_ohF1
z_jbnjl5a}pC55#s4StnOUcb|34?e9_G1l6Skrz54nQ0Wl=4-1`y~+-;_svFCb3esd
zR0Eej<^S>cQ|#femM+(Pb~MkU_;nQ3-w#)0Nl@&jtaCd@Z+xRhZ&0KAe_7EVxMne0
z^>519Xtl=15BYq$kPy9G6{wA_n6DVIRvo8Hlm?m@1z@2Y7MOlL$WK(YY+oaT@m%fJ
zu*LkgDqX<!;TR)txc$|s7|*af7%q6de5NvMW&9%ld<|vl*Czohw_=IyCHh`QasT_s
zwBJ5o)gzkD+^GFSt91nFs$bVIbCNc1pElCA(K$OAJ9UoK-cWUgbh=|K;8GAV0Wo2*
z$Pr7ApHd2YWV)a)P5o%Er-wZaHRaf!*Nq)(xijXo)fo1MX))(vr||rsDKZ`MUd&KT
zQA{l&MFeD=*ohrM#OcNyVPw?UZ(y!si_5xITmmpNbgiW0B1)`{p|>R_B})duqI;kW
zcU7=G{OJ}TPg_rJOq)y<X<W8id8fV}7xUv(-}Xa!*w?>x8{LcEjr-#-CNVodofjC{
zeKA)VjD;91>@r)rwh%)0@FTr=j7l9OK%!t3%TXy^FW;?XuU-ckbbGuXX%)&`63X+m
z+Lqdh47he4jy1#&b2h97*Q|`ZMyWFX@>{7Nh*F$T%2XW%)o${s+QuW)w>xA;fff#B
z5s~F-!sB`N^Mt1E#8&h>Lluo^j}HpRkhN!iZp-y~T0tAyCv7Y=T{g)b^{(5|rK=iZ
zoOvuu8@Oa{DL21&tWo*4*{@#}un!qDrxhKG+m88J7z<P6ir?h^A9cA8JEGLi3hcUa
zzVi{C!c6y2+g!`JsTyE0np`;?$YQ(Q%ADwK|0W9LK3cZ{E}<)?=+POlb)OX~E~PcD
zd{00bQYNbyH<<YI!pT82P1+6Jl?EMquwu!3%s!qz3^;EwsY~BQcsY+`exXNGVu2gH
zT`~{fzD)UXKyO8*LT2EC*QTK8Um2VsVa3~5M@3FpdXIY+88sMs8SEO|V#*=KgtBN2
z9KpG)HhJYtgwojT0_ntgk24GC&cW2squWl;olecE?!DDyUit7d+)LEg$t8Yo>eHjr
z<mBItj||>1!CTO$#K9r;)M$YAi_r|m;%BfVV&Ab=v8H~s3%D4HDb2UdgtyqG_FG>R
z&4Y9w>n_Lg6%1#iUV*8qMD<kO?Zv>wn;!CGj%1ZKg}7DoM(U71zr%S~gw2M!$cLg(
zpO&j$2xB3slDD?WES25+{ak{7R2xA}!{J)}&%4WSPw8q<`(Q>yT26Y*CQ*dPF`iJp
z%?pdoR0yO~Uq5E%kE5!<u2CT>u2rEOJ)SbvG_Uv^nK>})eM|&RXRjdyRVn2Im2^i&
zn#>t{*`+=Qz6l-(IGjx0{W-Wz$d4I({S5Y1lw;p)lHI>?Ez0f32><kwY-1-6_}U79
zva&-&_b89<y-62m&wl<(=>e;QqiT1P>)`d%Cg~I@Ye}h$2wm!o*wAzhoivAr9Pe8`
zm)KR<VL_K-*cf}-_@s|Wk*nXX-dDU2J?WkDi1}3~;zwSaNPvaVN4+S$2nB=i7R5NZ
zxB)0^i+92z#A)3*NS-BLDSqz8ieZMB*nrqQ!Xq}-B>9H@@s*5vH>NU}k<UlAPJ>N8
zf?7YjqkPWz&<b4$`|gZmyeGx=TBEF3MQUQJm)&_RYtG62AU(OMmwc}J41ASfX6@)*
za)i(Ej@(_@d0Kyrr~ToP*epU6+QQSx5cvJis0b+a_l>;O`b}cKLOsMOr9Ab!<ZHe9
z!^Di9<>jH5ztxhU?0<--c}}PLwyRE*d<7610Ge1+_ffvu;?{Gl{_|Fi@&OiUl3}@R
z46i!(Jc)T8{Hls(xLa-E>%~?X+nkoqnp~6nXX37~&bvZ8yIYfofs-Lloli$=6vq7p
z1E>=zrjVJof{oJM{yW}Vc4OU5>klM{8_Z_mk)G_@4fAw~XJ%MdY|%#0B;QP*8xZ`$
zCM-0<$-dE>?n#}!kmQw|XXHBfts(OsXijFj#-P1h27I=F7vT%-Zo}<R$M%e^61u{<
zipJQqMvGh+=8d$k?3=RAueQ3JLFMofv)rdn@C*g+KWK@@-TNn|p1$h?22SA2SU3+{
zAkwLP@z(KACR`9o8L8HWIT*)&6kCyzDw`J53D|X<7UBx~pxjy@Jw>RVPnu56i+=yT
zKj|ptB|Cs)=itqUFt$FzqX}#~V*ic4%-aEl#p|PQKGYd7yuTq5TN(C7{jyNgV>r#H
zA;p-Do0t%0BeZZ7bEu&A*pxhgwK%f#IaM;i7na4qO_5^G@Fl-4Zjs>=8;2w0lHN<R
z*pjqvf^&+IE1?TTu4Ge<QblOMah2@MrjQ581P6eC36m*8-*35byQQ^c!n44o?4Nwh
zm+a@Sw{O!UP((`)Hg9a#d#_ZD?1a;Djb<*<TP(Zl?AIgK?gPHZMs0OB;#Q1GXxeMt
zORLv@K=?ip?7h*U?1M{MnyZsB%veSkIs={F>syKvZ<&s^uY{*wY>}CY;<Nkm{I3LE
zr{c+fJt<0lrk9zCQgN>l6SjTzkY*&23B99EE3WjXzM9Y5NZCj$o+i_wEvkgQ(Ug={
zuR?t7@d+vxd!;=ZVLX*~x)pLz#y-N}>OoA{_^5!ZBb%sxI@(2zST`F93)#)wq#uSL
zB2t`cxS*XvV(7-*n)9g4nC2sy*{L9s99|}br~RfI){1ovGmj7C75=g^tJoi};on#f
z(|tEa@IdDHGZ^cW4W1c(HMnVDiMu7nEjC1`pwmIH={I3NSBbOW%_-JEpKX<E&T=Il
zDaF*u*`B=2mw65A!WYoNP`PYHlWS9=3@IXSU$7#%oyCj7vPQtyO3ugo$FFQ*L)DrV
zW^7%WIXL${)KYoe4XzB?)Usz7`WhaTS5!AY=AfKYjbDi-ckE1W6t68Ewf2mBP{q5-
z{)~Ewd2rZywmhzM!Nm*Dk!Vv(xyj&=DYg=k;f3vC_5Xdk$oC1${v*XhMpnqS1nYE5
zu|1{9@7=;&uNfA+h`to9R*IYx!Q=89(f!VFFNvRO(*4U2XqExXif*RF&cbACgH?}F
ztHlMy=nC7a*6sdhyH72gnN@kA)%7q;l~aM=n31kpSTt;m_(iTOM8{%zehXpj;K&X;
zNLy~<QS`++Mvy~dmB`T#C8Dd_UU~3>ZgKyT658D8fFMdKT{5+<M*7upEU5{VM^RBF
zE!LOSJpw#gl0#P_HSBVaqU61d?%b7~8l{Tom%o?C4uLhP0mySPs=`4*d<()(HV)6E
zNbNby&NHBZLcVA?jf3*T&zWpMyumR8-_zl4pYQKgB*g<%T2QC54@b7wIjnz0Y1SWn
zwV0dw5R+;W;4%u}8E6miPW09lT9~c~DiMmD^z>5czujnybT#0b4@7HnK~=B^bo53C
zn8HF(kl5uq1`6_z-Mjgy^PZu4L1Dk|v2{fJK!j!4=3K?X;>Ejw!E&AK)(^yT#oh{j
z{!%p?e&U)KgIF3t*oyP-YUA^pBW&6ZccD4BPOSgzyx@<NA8%d5A(5S}I)+YX#f(UX
z`#@g<1p{k?6hj?<4QxI=ti@_cx77@z0Vi5+IXgGY?F}7Hbm{wb*jNPZ>ixL3_{#r%
zexZSgVIKPC+k+^t5QVL{Rg&@Bt-{U0W`w)2v6zFR93ONxaU?-c$lb$$w7IyN)HjST
zwnFu!h7iBh8v;j}V=UPb+YfWv`Oo$nw@-OirpT&}e4xTrBNRyipvm(`uLTYs%4jq(
zFlcYXTnCii#9RZRhU_i0XLPbqk@8T7=`P7=ALmnFV1$ArZ;_x<#;tyJQjB3_;YH#m
zmrJwyX!4=48~%-Nnf_Wx^_)LmeXFdqvDC_ct+i}C$|GbmuW)5_{$`IO;#B1qYrk`I
z58;IO<K*+5W+7MSJfQPwEl-JWdj@Y<&b;|~AzS;1eStWlq*!W$lE*)dhzo(W2+XT2
z7Awy;*?N<_%ior770Nwo)SL=^_Qj@rPE%vo=-X1>)3$yf>(1F3%c6j#_nG6xwz5l8
zs{w@om9~8+N3SL?tbHmOo*v2^<$;?Nn}3j5kgpJSgBl}!%O$O6RHR2`=7y#FRoDLP
zgSKQZ4{O2S`+U7s)b^CVK+^rUJ}NQm+>#N_;r17<#CsGSj9<L^G$9Z>&QAUU_)Ht6
z9us|gyn-2}N(tN~{ge`zMdSCqU|MOByLv+Y;jmdQGE&+ZAjub6d^lph{wQX7G!GV8
zLdRSF{1If`N#z811EITeG+do~ChezqOi!Zyf!AN>!I<&Ru9fp+TYhAneE9v*jMB0Q
zIz>!{qbuZV?IoLY@9jrUN>)zSLx%$bSQDGjL0Pg#YUsa_IGn~Y9bm`aB#%E+qt_7j
z`uuY1c~5*)#N+r&@j&*coWI0=im8iToqqc;vNQM@zP3X#n>#4){6<AQ1|P%}&-ATW
zEE_KPYQ58jO31MklUcuR!qJ0`?2unijP|WwP0rB!nUaTM$c2f?h#_IIfOKoaM@;^N
zd_FIcpm!JbbkT-Sd;+8Ee-4Lth$+%*8dRQitKuFs>=I*td}HUEfDIZ@ZzR?7ZOgJL
zs*?j?KqvvR!zmvoH)ntv2SqY1r{YalDzbf_oJQsUWXfgW_i7(Dcrt|KDdw4sXy-8a
zYq+h0F+3aE3!an%Cnv)j)*Y8gF*KGqH<OpK(P8yrRo65RNl2^mniqUjO%lm`FY(4V
z`_gZ5;`vay#fn)kVm_5?phcB@lYtLaAV3T*Zib^#YHhO?0xpbO=8iL|)uPdxn+25v
zE7Tru!iV~i51sspcYO&IfBZ1hjphwBvMFO0e$N&{zWFxI26rE>z)e%__HMQ=SGEnq
z012Euri5<wiuc9crc}E%*!!<_!wI&y{Qifj&6S~so#qcn&FEp}m?-6WK<)*CQ1sQd
z^#%K>guA2wMxr#2Xk>9HY80P=Xp~I#67w%%ni;Bz`MKMD@6KO(`laLrkZz3?%UNKG
z&r@nUiX|u?Wr?3u*iv~9M;-^bZPrW*of1j4>>0ba>uv<J{zgFF^K2NIa5eLV!|upv
zH_GJ|4dYo_=tSY?1Vbwh)FWgT%oVLGB%b=Lte>4Y6q@er=&qAW#dr<S1^4=u4!p`E
zD#(Jyof-cH-h#sN%;M&=!~&1nL+%-UU&0mHnP(`A;**LkU%xGV9yJTu$u(cg`{YXa
zDC7Dx5;QeeC`CmaxiZWgNVs7N(km3@ECpO7XOJoM#MjM=cRs!KLI!GGC$L@3VGpO0
z(lMVPJmUlN-8kHC@-eI9Dr9hio-dUQE?B~9i=m5GC7$&?4O$TTg>4dvX%B90!MD6l
z4q-~@Yz*rI%C_wYTd|K+aE^(sOs6V+OHn*2w+7XEx&HP5m`LxRvT&|p8RqU^1kFll
zaPAZIoz~LN2Sn=OIzv=^65G1@8Tb|nE(YHRe<Fq-wLTg}U}J@C&=UF33Hrf=TIMDr
z6)~vTJ3`|`OPk|THZANb0opPE4q`E}$b2GjDk;F&dd`@2t^B8x8z$!$)l@mJDj#wy
z*AgKXx-QYYO0e@L-F>@711{m@sh0jU$LVwI`g0>8R4ZEI{Rh<7C}-WN@<b-JrdB_S
zxXs!iny9j`pq2DtWp0_KdBkRovn1c965zb-Jx63swZSM<v|P5pY?abmPMrok`7`PX
z#36RrjvE|EY05PSWHTEyRY!!SU(yPtq}1uRi&*VH?NSG`p6V@z&<{g=-2Yg0z_dp{
zU9)8Gz=g`y*sSustUnWid&a;JzlVzYG=9U6?Fm4!Td`LIUVK*!BROdT-JZVG?Ti)h
zqSR%pv#b$2icg~y$wvFyTzfntSiY@G*7t-$p@Xc1W}}22Ww*^9rNAe(^o*^)hvlBX
z=Vb?r+$dgOI!)7e<(=pE5{(*HbF@<{q0?ccx166#nm<TUL}64T4Z3*TL8C1@pzmp$
zDpKD|-PQHG(GdC|<RgzN&vYGht)_AWi7@u+eq7~X-+xDA>n<xQuPD)+cWxlGyRhF}
zHiJm^wwP&~$KM*6qE+G>ILb%5aap9GC}3ZL?${G{iagm1=SuCWxrZK2KWcT<C*APz
zD)Et?I&!xUFYK~UQ{|+@V5pX3(f(7lQ>{fBTcWDw>ND|C%bX+E`eQTKBPRhx0qnHl
zPdv43I=twW&*NRE$!^4~ZR5u@bwroD%pH#T3!XrF%^~i0HKf(s+R=0=yPm^c#&%^1
z|8*4XchMX+a+xGgM~%_PsL>D`7S41ot}nPDEW9~bL8~X@$+nb7$<sH#r++;f6luk_
zx8NPcn8ga~+@w0M=_OZ&g{{gQFkOPD@Kkdto{W#ydwKqOdrFIb^P{2Jn<*yh>=!wY
z#J&-nro}MMk)6?cL7gE?Az_>b41<X~67EfTsi7U+{MgHA*0t02cET}hHP2K0Bbg4m
z!?XJPw;4c-=&hq|u|_I>_KrIKi3Yn+JVJbBK@sz3D6hALz+j?|PpQ^x7P*tTpi28_
zljv6;M*MqiU6>0fR8jD}_&V#gxO{cKpkLxuoyC>H_liN#&JX3a4VStZQuqRniTZ(u
z?PAShyVRDvH1qVsO>5OPf}QS+^S9gvD)JpkTi|>F@+|2tU-<~omT$dO%H&;=@~r`e
zyIhMUeweHS{1y3frc^qk)2QtC@oPmRUDe5m%0u!+T-Z+-DVD%I{6Y=~5LiYl)qy%Z
zdVXsAgToQ*Zl_`o!Y&{<JUM%mE*aTGdc>A?>hiDmn=O4P97;46i!5*3(^2tb!(V7J
z-CvM*Fotdo=_J!n_`2p(hn3sBNqq%!5?>XFZ2N2(3%@Ou5!~Ck8QrGQwv6OQKCRpW
z{a_ps{st8z4bZ8dgSerd!hkQ!rPmxb;||J8%A+`EbOVc(w<TrKrS3Dav?qa!z~{$o
z>EB^x58Q3SH;%dMA4YL665*J(`2Nw7jmC}qL-R^{`@%`TvZ?L6OqyfGqHl+(RhZFI
zQFOON0K7c0AwHwwQD9rJz27`)jVLQOTu@AXpt{Us{~g^jPwhrKTg<q+l}ZYId`4<j
zM)*0&Zkc))!?#8p;lxL7)TnWlH2c=z?OrHJUm#o-g>n4BWkY>K1I)yJI8@}1On<eK
zb~?(O^5oZ^^dmSSKW)juzr!&tCi1;p61ua@)LsWO6zr`~tH?4^wu0hw9p056ZgZP2
zW{3Xvn7gBx$$h@9>3BI=`xRcL+JY!WsIBqc!ZZ6kisyzS=;XBDe-=JN(_{LnDLB34
z2nqYp8Ln5W_o~N0%s|NyXn51s&UX4?KdkPY!8HS!w$`dn`WGTli&EjfVwy7D!R(K9
zg+6u`tN>UEj-?`XFARO&OQwBGxdSLT5LROscLp_uw1;ncZ*#F1gX^MveXdJzAHgl&
zrclBG-y}a~hP|VvD7_t&h!42x&g*F=h7W@8hviz5Zd!m&8ANWreqr!@$TScnyjfH^
z9xueM%7g#DT74}W%6*+_qya0sSIQZW2QQ>te1|C0r8uqUTQ3#Kh$RvXUk;kLK&KLH
z1c1Fh8RrV3u%da)Q}-ICQ|^`8(z-(oFNW!pe9BvM0G}HZ6YIHI`>!vGeHCwCVAW30
z+bu?Jap%1EDoz-DV>D;s5;&%k=WGxyC-U>ga9NL>`e;nUORw<^z_$W%c3BS$x!$XH
zm~3hj{oEN@M2!=_{xm~8bK@t``Xz?NU29B*aiOZUq&lB|tmTQtww40-biN$kdayO6
zKbmShwq#*%!aZ7EBKyT7*Y|U(`-flTC{)x&(C$*t*}SfhyymZ19_STmRJK39RKh-#
zIU4FvWnEG5%oUl@mUOK&v!u@E-pu%(&E-Qyvx?Vp!WXn&6w**Dz5L``8ikZxxdxvs
z6;3CQN-e1Y>m_-rrr#Gx(3Q+e1(qmNG}mdWV`M>vQy^yUmy{JHw_zO`5(Sw@pfm#5
znQoI^)vzXBD0#EHO0rRtF!Zjag?PNb|2(ghz|3omi_O-<nQlj_>*iDj>`0~*B=oUV
zb>xXG?P(pCBK5_u@3%9-2$x3pzjiYL!WC6eyZP?Lbw|nW8dvi`Zf{{XRD(3=1ul5n
z89C36`qDr~1T48)GPQ@yLegB$Y>64mkhY>(pl#NC5njT9?WB?{X0ICz;~dkM<=f&*
z%W7>cs5fY7$&qr4s`Fo@>0KQ1{lmF;33U9ZkdlycvHB|ibPyA^Gc4j+Jj>*s!M)7X
zyj8AiVb4nbo*K(*5oyoE=h+(o4OI-!(i&A$l*HRp#F<&8=AMZsGbQPTK3h-7?Cj92
zF^t8^>)md%QZ_gk{F(Q}Mw|C)r^0<j6CQj>DKp47Yux|obm7l(*0t8W4(p_&gPtQ8
zU4%9H?%$aPzYlLUoi*%;K+Sf92QvFW%mr~`iwomYx^Opug{FdK#XJ@I`N#ij!}Q`n
ztKNI_%qRtV)r4ffV|V6c2y?NO^ScYtu!_vXKhn)#%abdi&W`5G;IEbaPy2auI4tDC
z96hL-N4qxThPUcj5{B|JtBZ%Xy<U9m`f8M-7~VHbiWJw%GIk7#_3CZNZQP7<H}LQC
za($`BewsLcOr(#>5Bu?(H4H#p(O3azlt?#E%W1ELUmsa=-J)cndHl_Xn_zH2T0XHI
z&~=u63=}IfFCj$LyiDz@iN4n)*?o_fCW#5X)@fe*#G_Jg!O{TvfP|xzfVgH5g6+hS
zgau)~sm-{6|HIvT#<Si2ao-)zqQh*JTD3x3vsEiWjZ!sIom5M0N~l^PMr#zcs?-*<
z($=n7f+D2$CRS`Q62#uQesTWK|M|M_2lu1<8m}u4T9OBH{PN3jeAnmw;rP=l_ZS!=
zc0>d@&zbo$HtF&cB>cj+*Cd-7jV?x%1$lz8#j>_7)-Yo`@}ZnNTq68^=V3(nP1c#x
zxFQ7~Xn?z1kc}+iuPI)>Zuk}_O0?dSn-VIZYk6BdxD#Se)z!DT9rYLZxb6)P*2s)h
ztxGYg8y_`At<RXr%ida?hqQF0RJ|0KA93=2b#R4O>DC#qIka_d#nvm--ne^5I<dq0
zXfV6u<nRmdvgqCur9Cg)Rt9(2ir1Nyp?3<zj$Pd5XV2^Fc#Pwvku^!_tqp3k8;|X0
z=QJN>UmuA*aRm_T9~Zu|;%S(G$+Q<cNv|aI*j}_H#mDEyx9yAz)&(BLkA|IF<h#$H
z(j@pshtB=7Wt6g0v!I&1&SOaskj$Ze#!E)f&%dLaoT5<M>l?gB4ZiV5t-iYwRW1i4
zpnERr_XlC3$J5sw&}0!57ZkE*!+D4@Ve8gg%!h3x?SQmFo9Ze&BA_pzRnXQj6kagw
zDzK=q_Nj)x<|mK`-37)z>`rx22Q)q8CHX$Wmah|`Jd0XyqLkl%m=~ySRM^u)BxW(D
zrx*z3yO}VO?tl7>ycv3{XJc{yZq)OA)?73L(9*D`>?%_Tt|UhzEagC%(%D-}*{1mH
z?O>Ia;!I5%K5PgTNb$#&-P$z<JJ1e~$XkTHoPNq>yL!=ib<3R7{uv^os759)Ts4*1
zc9ibn(=vKXQ*1}d-XNvM9K5&$semIND$lynq4L$Pau!}9-Gv+rxI@$IpS-X=e^Bs3
zt*vW$xLwW|GT74k>plF&%c-auzIx`Nl3R#pvS!gYBfVnuM?1adZdJK<91K$SOgd%%
z8n}tMVw@IWu*)%`ev4F5eE4Im&pNrvEwHhS>)Q%DYR@&Mm+kuW3osChzWdasy+tK%
zpk0RJd1~b|q16gr9UnXhzsMUTM_&Z~%scpfi+;uqsfNtQNGdO3R3mgj{B%>WGxLa)
z3|nW=-oSRpU-r6SFs+??L45>SNOa_4hxM@j{?byoJ)`+YMW1Eo)TE<c-#zHk@YYsD
z?QT+S0?hE&HfR_FXE(Ya7u@8t9>y#KHjZmX0@9g;<H+<Q{m15QrrNUGDjO+-PPvym
ztf%)C1EqC?V>*lQf`;*%Ew`Akf~_=hcT@_8zxe;QnaWPt5&}trI%2C5AW6~;5?A<x
zl&-`}o>f~dT;VMMk*xi;yCZ#E_F8KXS{Dj{3Fd65_u`{soY~a<dwdbNxSr45Fc0Nr
zabi8j4gOamj>TLtH`w~3zG<-aVn%04RddPq(&k}A7X!$bAzC?zp5Ih$1SbI_uMVve
zAJ*_mGTRTTrIai&l@F*Xf2Te9PUMDS!RgJ1nr3~TTjkaM<)61=C^hQfB$suFnPY&p
z>Io<`gD2gb`S3T0{vOc-?#a@D6O@{yax4gn#xSX?HIns<0&c}QQ-jNCj&uThaIp?~
zqKnh0hXk?A-V{N&HePheJV4Qgk)noRo(ryT$}SUlQ>bs_i{0ThM*|bJH#nW9)hJAA
zLuu=L-5=mpay2o5)yvC3<o6?VUSif-03uR1m(uXTeIh?iw{K|4?(W#w&YS5*`5uG}
z1QK1l_N73FvO?LX=uuV~DUyeg6tc1vU+?Zm`1jh<_4%CAup+70swLX#DS(HO1awLz
zbSR5}`eN71zCeROe|Z1h?8&^;GS}syFdI!$n#3UI=}_|~gejLDHjwhQ6z|R%0y&h~
zRUGB6a_{vUlR&a=ooV`UFvesKH945NOyIocaJ1L4+q07Ngcq}=csK&gY53|<XE87i
zk^yq8Z&}S)&G}P8YQMVEWEWvzuX@BF1^lNng;wj0H0&w5{PV(qC~8;I_9lmLR^kD<
zM&M?`0qePKK~QmvJVc6K6tvS~OFuuQW-tS#yUHVa_Cr+*la%gWiI5`)NmxRgb`Tu=
z8x+3Q?_gZIuxw_gyL0vYJ<WuzfH%b)UD+6UZTBILsE4)RbCj%7X^QUdKlUxV`^GvI
za*AFnK*6~7ShB<Cf+Nr>@6;Je^XKEV^(jo<xTYWJibHhF$XB-Zt8~(!@N=ZQa-iSL
zx7;k^FznM|UGo|tPLec_%`Uu;`Lqmb(5S~_<=es;W+%jKLh{$&*Sv&EkMK10mto3Y
z-n|l_Q$L7H?P>puAWS&7zxn*f5w)Vq_PI%|GPuG%*``vA**#fpBXHG0L7wmvCZlAZ
zlA;}YRTS-n>7d893I>=to{b};`o)`=N3Lwmxcel7!@uz5f>G?Dz${4-wGe_)GN%gx
z;c1E{IVy^=t2CpR>n}z5zcRh4b8wZ*aIMMw&&-F4I&FvU^?)wTHo^PM+`*55*-mxP
z4(8ha`NWDOxdgaT1Z)Zlh;LtCjYYGO9mPORKy<*t*usZqxw65?93F&=1UkhFsYeT`
z*}ex@-{im+KDPzsNw-V0gT82+YF<BXK8=xF_iyD#pcgDFZklO^>eZJO6Z_1I#PO|;
ziX$W4=?;q}!uz(?9J8PvrbW?d?o-uDR={ZrtHAlR<fM2Au7&lA*I(G5@z711zMyqy
z^WCK>kvcsC9#`U{(I={iMa{$eg<Z&4=sIL3(59fa$XZa}#kvXQHh9#Opd@uvqO6ah
zG$})iHi<^V1+&TW4#wxflXdq6Wsf$*PWJ7#df0@8r`&~4e}-3~(90$7oxiVNM+k^8
ziw$LLT``@JC$3R(NcP1JS1$|<+~p#$*wRbD%+;<eihC2xdA^}lf+kkb$~=WAMe}LX
z9AfNPf%PC*;h=*UY8Ti@{*4@6eWT}16(prTJ3J~@Bb=}OUS<dk>4XbWXP>q>)K!pR
zG+fl0dQ4HA;~1v=o55r!N7VpHLq0Z-KktP`Y0gJmGwSgfv&aUY08-OvNn}{S16Bnf
z&3F^dAMKzTOdsXH%@P&x?0bm(3Q!cOX)d2ds<q)*MNlD$Yw^iLh;Ypqp<f^eP~@7l
z{`77(9*w6fZYj+<FCj|Hr^XX-ruBgs$BUI6s`}|a3M>k#77z&xOu;UKM$w#<5#N2s
zrQz|i2+9&=uQr=PpzQiu6&d)FU?tc19uCnv2UxD^fPNS~*LiGa9SZ3GwI)obX|6`Q
z_$9A4WJct{q4Cy)dOB^JiNAiK6u)fW&Q!!2M?)Aw8ZnaQ8v8azBd2jU2KQ5#Pn6a=
z{q86^WHLHjS*1J=5RNA<%SXRi0-aAThOa6_dh{1ypt7oP#pV3qYdn+IbTMF6#eO>k
zAnL!{C0Jo+j>z!YZ8br5{>6YuQBYnjbLMg4aaQ7<id$p)a(QV+u{-Wi!M)g&FnU*F
zAgW+vV~Obry`9hLEN;$7@<cgyXTKKMI}tBO@c7A-&wNmuBQ*gpnei!5Y>$tuUdV#a
z4VHeO?{h5r!db^}ung$vHNEY5Dc{7C0ID2Ggsf=I6jfNjA?gh@$jc}p5)DoL*N~2k
z6K9>o+y#ac&nM6^lY_NJSh#d%bUQ*=K;#~#)Dr^ZiXrs1KbdIg6c*K(4yn^RwQGeB
zWm#`;4jj>Z8B3`6UR%j5ToSz<aE9k5*jXf$UXb~9=G8anJ8L(IjtJViwza0}YhA@q
zKj|oYVz;eV;a$8pM&Ohp!pw1L2(`Ah+OHh2d)LWF5hwT_xy_HTIEN{HqMEYtRmpON
zSkKNe?N2)?NB0I|Ro-uW+6?9V{>X`;&#fW$a780%z3q9FF7J{6eyF7wNrJ5uQm&>W
za32psW?H6v3osy=6%DK$-JaS3x9aD@j*~UpV>nrZv7i$zY{{wR?)g_NIqD0X>*XXw
z`E_9!Jl$@`Np)|+>4J`G^%wP!542aSVv){C&$ikezRI@T-{G<}4()6dif<XmMP=7?
zM;4JQm{kNfs#*sIib*8y*d(xoL$)((lE4qJa<PCpHhw&Kd<&@lQjP~3y+3H>Ykdz9
zQu(Bb$>)Mx)4f`D+hZ&`ETmGXN!{p_h{|t^Kw<U*^`LEZ$bjC2X@<asXzS<|zEciR
z$*73t&stRtlr<NTUZjSh&YJEXJrZ+Em60)L((<l_<h9vGZ7zNp;%md1^l|P?SZQSC
zVz2%<jJsac)1JftM8{=)`905xvXoXkcl$@leDTB!eeZ)UG92k4MWLUXn&4MhLqr)t
z*#*LqDhMWeuA0k(MaH=IN)Jq;Jtk#Z#J4Xks<{ZBzp_GXWEVyc34CF)PLP(Uo9r(|
z{zeCHIWbJ!)@~D2u&X7f$=0V7%^Gw%!@{71T2n_bA~+MO@+W$G)#fp@tii71rZ2v2
zdS``lSlelIA}q-6y+NzM6P%;hH0Qc|k2!0D11`&@`gPT2SJR@-LAOjtYEy?NVbjUS
ztii66^=Ml@YxV^Ra=VGll$@0(mQ~>0uUSDFBK>(b#<uT<IOIYwklM0whs(*4o7Fpe
zP6Dr#fkdVNCWQM!E>-N)@<9Jov@N_lZE0jK#LsUGrp{g#bxK~FRMC1-oJIECOFcrp
zR1gg+@%^gRtpU{tp@DpZ-}4404^tbJquCfGCYi086$<2|i7?%gCiUT&5lrf;a5y(#
z?m@^S{K|zA6uq*F>OM>7#w9-8B-Ted8*~?Ff>OR?`G<jKpqS2y{_uBZEESo}(@0e%
zv|~x0l78&n$-p<NxFc#blus>{9EY~OwY#zuE1wf*7@Mo#Zxv?^t-0@)v)t*&`h1`P
z7tjQd=v{yjR&yMxY=~$06E{)Jkyzbas4b(1dqrqEt)DP>v-8f>$IcHY%8_r%*<d%s
z@O?nT#oqn`^yR{WvazGu8!mQP2R-=jmgVGIlr2Y!8yhTWRn@*bsAcBKuf8vXW5a4-
z(T`wz|J7SHtj|1K3ScEcE<+zy!b>;VNu~z~XqH9Jh(hEt-K;<tOIC7RvULo$Jt?Zp
zxpbqnpb#wZ8~}m*no^M2a~D1eaF&u7hx+dI{+;bm_b~FH@8RIOV+pyExzKU{6or~<
z5WI`st&@1RW*|SS?SO*_PICDGupzwx_(I$nzHUtNiNYH}TFg!zg6^4DFUIdYmwCM)
z%=lpcRiciO&n|VY!8q|d&tkD6ogippD)C;@yMjw>1FBc~?#F*)wYc-*ji}yl-K`+u
zkJ9dUV+0|==2KQirwV9HJQ$m3U_|q5o+)P4l1>2o(cs^eE<UWLSYD15NyqrR(|JR<
zJ$jf82GoXPLRY5_J<q?_-Y-s{6qA0?M82ASywN314512SS+>4M#PS#G%KXr4`m5mO
zDaATHUlN_SXtG<ERjk#|Yu?3T8($7@4hfeGJ?DGi1%qI2A65v|*%trKF9k**Sj+A|
zFe*?C6!N00RaA+8$s)g*P$qeX=xR%OS!NYx%WF#ab=WP-x~+ADeKf@xB$a%E&g?72
zqXeKYrKW73#1Bj&GHM`WC*J2l>`zYYlT#`Sd8akx2<qm2yU~Asv6+85EWEHXG_ktD
z`V1_GGVIZ@_$fMm;TtI&K`J1@dwy2)y?ZS?ZFK0dzGng_*^Pvj`mNu&($e9E-NjV!
zsIJ_RJCt9{Of@;T7WJXhlj7OraTfuOGfXK&!W>s7zPB4+-nWBJxj(2a)~(eqYsYot
z|4bFidqaFohLoywsM9TS3Jug4urUj1V^&|469fmH<D3mQpRzr5KfbuHSA|QuS(aJg
z#A=xIubWyU%=Pi_&K0!LutqzoVyBLLX&9q@>!Z-R-{?+WeGwM&dG8!Jpr;7>T__J^
zp(Z`X3|d)xR?B-huh~mwHxO)9Yvp#0M>>XmV7zc{@9FBD#jm2Fw)B}i7#0|RZdvI}
z{6wM1j|d<3edjfd(7=yro8nR%WwWbN0e^}_a_>XOO7i>*(_enE%(MCK5GqrC18d_S
zZWbI0xkIAZuH})4yYgEuLOMl*Dc5NCEs~~+Z;-MNamePqdDd}T_pK$c8SDNd6Yn2l
zw8PKk4qIwsqN@u^^MF}JqY|=|OzGio>)|GLe*2%TjVrVQTvdXxg^C&Hi$+zxt-r3-
z_ttn<w)sJc3F;MgGtb%??ELK<e2`g1nH&ObQ=Kz8nl*7)-=T_VL%(()TFy)lJmsi~
z{;(1px0gz{0t90c=7YND4)b9jE)*p>`8*PzY>X~bIGa;VT(BAS1q5bD>=Ao;X3AIV
zuyd_EFq1UizPJfnUi~dG#wj8W`;?RY*0ABY?Rum0mksYuD{h(_-d4$BJE4<J{?oWh
z1<uqeI|5NBF+#udiNDwqYmm}oYpf?-ag|%G+y~iHWsUi`qeAyBb-C@4QX5p$Pp66P
z4xF2VP)8&u9nmhc1sI4w-hS)j-^9NY=TYJ$&Lg5-qG8d`k@rAkkO<@fspwA?Mi8zu
zF_3*bA}TwCK|%tEEQO-oq8%C8_<%vubB-)FpVdciPVz!cd6Ds;O^}Y7prjVaYdE#D
z#v60Wk_Ybr_V?1o!|ktyT2|7Wg{2)ZM9W53lHXi}nf7SxT;YisHWOXA74=AIlLhLn
z`MicT_0#X1c0{!5E{hbMb?EL7WxjweySLvvK**^)>^vA5HTAu9{v1DTQT=I|zGuJS
z9EP8BUv8giks_LNDH&O4n#LFz73m)8ZggL!=^{^j>t?j;HH2s+4);8?Q{@`zp(QSU
zypbOtc%5<%XeM0aixkma?XF#8iMu1+McI2_q4)czd)Bs6Z$T0VNh3z!ETSKzF6o!D
zXD5qg_qTV4Kx~3}Lh0;A!#H`HP)9$}c)3@dO{|@80^Q*f*95v)WQTMu><K}_@)acj
z3*)ZgJteD8Z%nUTV{!mYm#DM}*<5iX_Q`P>CwDjve|mYL)0x1uMK}P}M#XX}^{_$_
z_plY9s|$>_MMC@~CGgP2bE(?a#KD%D&F?<a9pUcYOD}Ffj6gk+ocHWsR^6y_E-d=J
z_4T`N@W3=0WxqG}^rHRfqc)G+&?+%~M<Wq|J}u57)|JcHqaSI$o1tzN<4stm3)@!S
zv!!r7F02{Hzz4&F4!O6>MjKOZ#%}rR#wi$b%KA6*N;1j)55+!NmqRm?0?fvl|5u1h
zG}kF?>h;uphgI)slzML}rWSm4HU3T#<ZLbSllfbfH@8Bh4KH<=ve@yTioO^95HAe^
zi7#@Cx21bw7@~LhRJ9r`-d3;us)$Ra&(})U%A}`c-216-!^!%LRX+nspBs{2_5>Qu
z9Q|2}`Ervx2(zfSF_CUMblX&odxKub2RP*O;Ui$3FwI1er#XMVm~P?hBx1H}8+5m2
zk!jpQi=t|@4!Qk((6L9K(VIWFqJB?@x0sV3(~%}dcXn#Y*Yc6LZ4%}htq~^xlibV{
zLaUpWowWo8k>cT*?R`(+wqTs!me5-p#w`LuMhFVKG0J{!O~~QTGenW2v2|>6al2{l
zS`U8jk$1_-IyJ+M@q17s=7%M53aqz9)j&KfE|{TvGq&fM-v*C1JpC}3TGpj89DX9>
zs75x{%PQB_eA6^FfYW;p3`<{L%oWoZ9#JTv)u;})o%ap;Y2!TGmD+Gd!^`7OD6Cj}
zDn}@1+Uri6q#m740QB$%ui43c4x|g%EI4ww-9X>C^+iaKUAJ2cFczaPEc3&w;pv$c
z6e(-zQd4-Lx#15t%g7!<E0Na1ZYgmMBc{%o0uoL}Kk@|DDm=W~dhg*cxY<weCVNKe
zGmq<$2OFj2`hlSNvU2@1mK@%{LCHp8l+6x8CQnnAlURNY@1dD!q;rT&x7BdU;d~C?
z8TqW~CrOdc0saQfWnrA86|PB8$i{4rm%W9(lw?m`M&1ohusg7z?zk3^?krdBh3RB#
z6N_5wo`8yJcXh8rVMhKO&v3L`z*|n?VZ)^JMvpY6pDPes2y2h@%>OtllV4PY-?Y6!
zkR!ZF4VPSc$$KwSS7FB7iylmT%RHWwS^3kQWZ!0&B(+@n#mDr9{h(u<T}b27h2R3-
zFZvvP<2gR7zY3@91?>%$!I$@6K9C4H;ys*zr0dyhSBH66_FeYb$9<_GkK~MDfAu0<
zIo(=fxrmR)n~f;)j5y!toO$kIm;T#g3<!yC4y2E3zQaGitX49~Y++lzB~9O5HmsxA
zl!XuIb+5O<N;y5za23o-WV%+{oT?Sk{HeJ+xV*YSy+iO};$C#H6ay)rtEca&`I=(O
z4`rPij6fEXcCc^H71l@Np*fJXhr+I}PJ^<SLe%XzRQ=Tqn`V-@ZFpI0rBry#Bqtvj
z(w*A-QXK}UqB)c3?py27loo8)vk2)Np?T{g(1Eg2F^t{edGdX_zcxa>N&}@a-1Q&c
z+A7QBUI^pdl<khCd<&8Uh8sbvv0>Vgc{h6RIc9cgV(9CyI$m|l=-lKRc?gA>wuui5
zBP886Yh?z+K`S+vjlD74etR7_lu~zgQ=xRYEo|`#=lw@j_aFJ%fMMTq_vr2GDh57)
z)&X|n#jWoveb2^XqD7`&e08J?2cNF`gi=`1WxhCkN6yl^cabkwhAlFWp~_7uCUvEh
zm-DvU*RRU-Hz3y4){YLZ!xW9MxTpxoEO5h>clH}rlxFnd?|9%y^Zvk$&!Q5ME|b)>
zf|bL5&m3ofDQ)3jUc{G#NSB1ju*-;xxe1B6L7E~v!%Y|Dq4%QmIAlYopYP5rtr{m1
zUnt)X3fGQn-`Byn2<e`h&xpH?69zLq8`<&FPv#;O3LG?v1Ubj5jTfKX>;d|cl{*fX
z)F0S}m&t<q265f7i+Nlv)lq*UPb^9*1u2G)D_B16bQ1XE-qG5U=Y4P_HoRX*PdIOG
zg*&+SY(zeY9@U<lgo^<-cIw7Cy%YD$4@h&MPGq~E);)K2))pzSmlvy6WlpjT%;0D&
zocxOZ(AlxvtlzVbDqRQ<A^b*M-UXQov}MqhvQ5qkB&!YEt&iHow5N-LoLe%95&B4G
zD={Tj>yqZ-GEQ=tyXchfLP7uwUTGtpu{|hclj*NO{iiVH03EzOFJ%WyINa7*F5SH&
zA8<=wFyh)sK0nm<{z!uh<ZB{bRt?A#IJQxsfSzZdN%5MUDMQGk<@Y5qFWI3&yo}*D
zqoku`ySZ7lMQY#XWbQHd=v@D5&QH4f?Wa6rWE9FJ#KA5w{BB9HOKCB!!L>+*L?WvN
z=5{l|+uZTlWLP=uYdI>ab4scywu}o%`rKp;z&x9A3f~AF!zlBKLz{APmjtYx%}b`K
zvgAv<bxrYY3?LcYT0E+#;S~{>v<j9u8t2AD*Sc|RJ|5V7(9|7k-uAIdj19YEv)%Z%
zC6pUr(LJdiDsZaH?jPKF<V&mH00*?adw4ni4@}K;*VoLVi3Z3_UX7uU8^cc#5lK*n
zjFuYJq_A;RH*KhQb+BilHmoipS)%V5grvCl$mq3^d1_vsVy%h+OzX`THiUqxc=l}9
zz;Mc{d`I5!Go#2Y?i~}j<xlr~bYt&w=(L^*(^Q`thvJEk249!gxWdqlgQ?Tt3{rmD
za))z^ATKGCYtNgEUbr>OwN-5}_nuO2Js7(40xysYAGu=EC0^fbq8#bN9){N3&8kM`
z(PhFmzwf<XaytDgd4HR2*n#|NsU=^xPd}yKNviEPg4Bw{HSdUV%0`t&N@k{D!v);1
z*~yN9BTp$n6&-0_E2b=+Q(mW|j`^&bJegpb0SVwWW&_oW{~0v<{5$6zHi-TMee|mM
z=%DJsBu8UaU1Q}NhUi~TQWA=$bjIStnx6G{c4rz-2gM78t%&VPs~DVnf5!*S6MRDa
zK=-X~n9P;g>HeGODz`JR+`N!CAVClkA{We9)oPZwXQ@M@YqeQa2Fm`^nx4{Rp*70)
z#ui*)$AQ#u`P0z0=DXq%8%CQz6b-fPvEGY(l{tR76~hzEGJJ3M!{Paukjr6857=xI
zeus8$;Vyv<N^B=*TlY25o5QO6Vslk&<YNN4$>HKN!fL=1Ge5b-#+rW3U0l9rybE|h
z+cEP2o7;ET#$iVW2^S@WIxrXY9DAwrhaMi$=+nnL{JLA|GEC*DferUn$q*8E9v5>o
zQ)(3n!JFIf-b&qVv(KC&ohMqCB;0I{p17J<Dv~K#|1mRnsh@eVS-d3E^XG->CkCh5
zUj;b(#Fmw<CXjPYl6zzBaxJW0T(9n=Gar5(Mmrvgl1Kpqs3M;*DcI;e&6?rZKTS_B
zx0aSa0J#nzhg*KtRa~i`+hXdh;4wEElkA|KK4@Y)h{z>nzTdJ4B6{PpiZD$dbj7y<
z*MS*&&O)(;g=a8Q`9tHy+|)2;TU*3J{<e`Y`e@3-cK_mtpj>A-Cp~z?dM#$4@b>IO
zC7F8TtY$W^rW;Y-BI8qzn>*p%1EI*=`mpiz(tFTag%Xq!^HXn4&(q_@r;WphO=^pD
zN%_fLAAG=HQP3DzjAOMkW{UkWTWSc_zmI+B9`2_b06Q4sz$NbOahqQbEnIQJ_t6r4
zyY)5$m3rBlKAD?@Ed&P--k|0YNpvWS%Pk->AkR59mElnxl3WPVG3LuPuP*F~&0h^&
zP&fm=?-&D(C$AUX@`%`&xxEHZ?0&^>ccd|dTE4I^9<zichK6mqgl&fggQp_R#WKS{
zZCkelq!S*jsZYK%XZdwQG}bo8E~@b+EN7f{iV<;TfldPN>7=;>D{V9b=f)L;5S{~c
zkh*i-0iG~#jpyJWSfo{-qf;*Fd?+kvOv_wMKsS1?H0*7Jc@ar+&jPJ5z|Rm>SH@kI
zUFa33>$CVMHBs^R2b-S)17!*57!M;4g#cIkf(Qeh%K|Vmo8kF#99!t7`wl)nfD+^_
z*2HUhWjxQa@0kLAMdSegsz|q*mo(#w2&)W(N4&2Lb8{wsbLio~4wp#wu#-w|!EU#d
z$1n4PA8C+-jfu&0i&p%Rdd72*m+k`{Qj4WFcA0va=ao2Xbu+!D_up3tT*$p_CAsGO
zYKCCIIQao4o(&$hg6%eKFU^KG{``23j2kR#DMYbbzHCESKZx5Tnq29*vJh?_)one^
zPGVzb=7b*i)>zD^?j20q%JNjyPtQ?X7HISvP2=Y|7*GChmc!`&t!$L_0exINIt`wK
zpt~sXAlm!T%vPSU8AK^MyXrN=vA|_-qJ{J>J7M4$N>vf#-0dj?FhC(~|5~(Kz{~TS
z|JO)BmdUz;4XDM%8Jv0@k^{Q}`5hv%SSL&DW$B^mGPoqVSv!|#XLT~><%l^BnI9Wn
zuz3*}FAw(CtZe~Q&e{=LW!^)gRWsfer<N%&F-ka&v#_;DGqZXf-so{6xcZE{rN<kL
zRZw2l0%mEnVIcUB=w-bnNVP7h*8F>!YV$LP3tA;Sr!3Y;6!mP}V4*JZ4P;8xSGa40
zQQArA=>^7LMTf!8lFWUN>eexUGWI_(c2TLV#PvQC)>U$eJ%8^~|Ngz|eqWNmg4s$Y
zDTIt%ZUYaoW*wTOVt%O5xE=}Z6uTq*%FY9vS+n(S|0qEIZYBgj+9ePU^&P87Wrj4H
z6Wlz8unSDb%JE}$c@axR7CB9Xab^;BkFHeMgeU#@wW$AxzC5Y_Z$<k5eChfBI+6ST
z34f{G$9J@TV7nmUnX!kZ=k9DA!5k1kCow=)e=oc{&4Ak@rPlZY^82}!@<2ISZkMhB
z$`R(?V9x0tKn;IL*r?v~jf#pJ)W2&Q;0~{q(fjYGl*@zAp%Ic(I@+1;r7C|H@Q9TZ
z2IYg*1;FSf8NU`=u0wTV;hKxm<VkR-nEY_8T1R?s4eb`+#jk#l)uD<JHzM58(@*$Z
zvJim#OFNbd%{1(bIRZ52kQC3w@4aL!uQE1p1>-@r6#4mQi7G}9>=cNEyj7Tuc{C@7
z<#KMznWb8{{Qc;+j0{QZgsiu@^y<p$QYoTV%o`Y1XtH?qM$baZiWfoDz%O_gAoDm@
zQl;e-Mli9^ck<fx(ZI{baxdRD3mwbQ&m~4GZsA0e@qpDyuNtFb$?cg`xv5}*Z08Pn
zYc}1GysPh&0FHojvXR3t%LwI(UF!SmR9)I7q$<ZzftP<jo+c^&6#n*g5ek^R5IuHB
z#*f_Ccv1n@pR&=GQf%q^uop0Irch0lep7?bXOUFBcQFcN>H>Gc1DC&-s*OMQZL0c^
z<d`a6G2ypRF&&Y#&KI$^<uw_2d2VC&29b(a!_Jbg@zx7NyJ&gfuh{N+Zv()Dn##vi
z81#SMNVFoq-*bnKs`U==O0uFVwkBmsWddk>jLU_lLF7DqA{r>dDK0%|tZ-4xAFZyL
zllOmbC>CvGgo-Er^~h8#vK9baWbZFmIeP9bpin`)K5+4%+0}rDr$jm*lHp%j?taxQ
zoRzA4`t4skrT(egA!%q45976b!RvdzPK;fpOk9EG<AfqC5BS{YSXAURfF(#|P~7P9
z7}~=PMq{&q{$QuVQtmOAfpLW-zo>FwUDxue->`chrRQo15|HXU8^_j!IuF~T+WEy!
zKHdWup8vWz?Ro%H5~~9<9$H>A9v)PcF2XyX)-)RR&2k>|;1sHY0h!G6ZpPW{M@cGh
za}D)Ir(J_F-r-MEp!)_<3CYHEGRHmpp~gNLl7ohlr0$;=O7oO}O&+7S{&O9Z1_k|C
zZu<f)nNID=kXU4;58zSMRzKs(Bug4=KHaO7O4*5|$Q>Sc6dLxaCJ@FQdhr)y4Xz&;
zsIJs$#kK%!0ISFcemQTMqyu@e!2)5c491!DH&R_fP-|3Am49yJ^1BcCGb{-J3S*xb
zX#1;jmO!I%YNqUJtY`3Nf6Z*6RkY04yX>y3Ou$#lnqW?px_{szdK1$%DZj`>_(eb_
z)ozas%N=e1k`BWCOY+ER1vUiTOP^MEsXjz|z~tR+gxRql$uZ2rc3=6hSGj%P&u7)2
zHyOW<+-}3i(O{{!`5xdrq$uG;xP4!GGv(_7)kDc;d(6EvW1eaeV)(dvAv-CXD#l{d
zQr`a6sSD7fY0Ku0go$c<d5+$8TN;~va7LBNqrD|mPJZQU<1-Y%OvuA0zUPj-BF7PR
zYD+ZM^LG!B0hP7S?jufZ&Lv!5`R8&34}y~~E_gqMt%$j4(+CmAoI4edSp&5IScMfw
zAE*8Y(DdxPo<*#;%XT8;c?>Lk^(NtOSX6w0g|sn>u3dEZ%kG>h%H{u-{lqatBJwFR
z=lB?+{!?HboJz$4qQA@zKN;n06c0_Kc>ugu;Lx7?y0o<*=r(-rah-P=lJ5R|2`#D`
zta&$(L!037-iH7lPuRY!Naa&#OH{kCsZ&u0^2^pr>j~Td+<yvopQiNkd|b`OSMTe7
zYi|Jdj4SBfj}`nP?!PN*r}GZebHgxews}tBSLw!tm}5?~WJY!E0F#6+#w@<ndwnX?
zl6cmh9xT<1PXi!I2-G?8idUY!NqFgJVN4?6g6mcD9L?Gr09gT_EThuD=E)PId@#vK
zfE-urQ=82(0`We;w>_S>L>jG0pGjr(5F~v!pZVVQRl4$-&U4Iluw8rYi~sg@(=V?o
znMHuPfETVmP`xYRdg1WA3i9vJp?IfEGfl*yhCSPXUeOr&9RvY2(q$?da(Zc2>l$Ro
zUbYl$a3<L~1u<0QEIvYN@)bRo&)u}^uJfnLE!gt6{Gb;s*%w>dkkP?XWN29Ovg6Cf
z&uS-9l{Wvkb6kv)YZtcxj5VP+&DyaOAelHFrAy+dzEVS+k0*~e1?sV~eI)pKPv41&
zdr(OQN!OwlU3iVBE}Ae;wMrUV@1sDpgOWs}K0G7H*6z0ddO@Sxx<=BvPqX<Z3+l5L
z-Amo$vHH5H*Y=O4hPO_MoOFj7Ze464)#<M(0w)>44ab>sh5X#<rYFEfu{t>pR|NPh
zbLZ-V+atzL%g)26z~jSL8Y5q0tgx=(Uj!UZRCPFwS@i1wzuzuf`b(BFRAgsQz&jwD
ze!k*Y;u*E`SvqKaFJxA%_8XN@vE?dmRj%e9f+Kb{zW{gfya00nU!x=_&69hr1HNm~
zscrn<sw`ta-<Yo$nc)A!82sLP%44lrxRX(bWtqygkq8#mFZc3}qB3ZVhxITU+}*u_
zF=!v4ozkH_6OyL>%+@Z~<Y+CfmuiXPk;W?R@zwJuWJTXBcXm0pzjjv4F+sSgX1eZv
z)|UvmfZp)e^8iusL9Q3QR?+!F!|5-V%Zio)O7vyV@ox$G#2qN@aKEi_)$BfA@NLGE
zF^82;->!#zx>$6+f;?0CK3*MNem1usS|nO%Buk;!;zHmYWz)?#uKFoZdbrW3(86)`
zE`u{R7xKAn7!i+X6pwM1hYaa)S6)m{9RL87buXr!ru42e{B;3401K3z2^8{e`6-j`
zlcxq#g6X~fpD@Dkk2!qOVT#IuteSWbJ?1W>A%A6yf7c?m<=u9*wC;zO??M{<`yV2Z
z*^8@X<<+o`pa)g@!t|Ew{)<7bqnB3vBx6s?*b#quOqhLQN+dPiymRmEva4oPdjzK|
za}xKU0WtQlQrysy`YATvE?+t`h;z78G!B}iu4?^ZkcYs2yOU7t$w7vW+I()b``B+E
zBL?2q6M1!|r%82Q?dW!Sc;=80b6;cF`^Ms?-my-leWEh>;t|yv?k_FvZVBLQ-~CQL
zlNpM*FCjteK1>(D<6|+dgG{^LsTM6Paw-0ft?CVQVQdKVgOCh}hIhUwM4LW+4TKAu
z{e*eo1gBwsH9P4$%Zy#y2*2pYqGE0cOm$cSlTxu?fXizIJdyCv6A9ize#<@mA$1lH
zwUPzrol1=nxFA2943>b}BGg$4{rXH=q21&`)S;5%`R3@k3&z^L^eFV53&s~-$yd{g
z8s68#l_Zb#j(fyh_ursHt+^D<(zg142`*R@{Nq`Vwp>T6d?lD_8MC_n6I<{<O?^4c
z>S8@HWNQmP867Vl`=?>-wrZ9nT5ha}-93yxks4VXoZ#bMszB7`1PP;NNx(t;QAcoA
ztR=^poGw<wx&NAc<OSINS*_@wx<-DoNAm#rgOkMFhfFKVEgaec7K^!XyH<ikf2Fml
z!p*ccKYf^0IyQv*!iTXkVH;6_k{IeQfdhpSTQp<S{NH$)qq`<=FS7e(=X$M3-s4`a
zKIdavEza=VDZnT!sBxo_!{e2aQx<p%UdwJR01SZt_i{e{=W>1+1Nk-GR3%Vfv@{7x
zLwLDVOz<VH_H79K6p{8^xiwDSD$&8G*9#mBpS%&#{H~ceET_(~n|J9}`6VnOHSdz{
zzp;ty0-MUevGe~f?AHGY8Oi?-{IUH*CxGsQ<@r0e0P>jyYP_QdMA-`^y?RLu@-v_E
z#x?AXal{5jWr!jGoa!|V;+8pf1WXU&soWe1Zf`w?fl+~7RBl$PBQ@Y(0K@^)1C9bR
z5)!osYa2nz2c~7?R3O-YDP2qDfTITr>G`_G0S;tM+0;zr_#Ea2&|^lm;kePl5iAeF
z-t&NrMfo!**p!*LaTN^#h_h#N>k?yi`<W36ugk^{w>t-6?j8%X&$6ve;7*=%yJFZ_
zkH%1mYV>UHVYWZRsIO;39j$H<@7vYb!PBVI<u7tz()kxs0oTLb;Vv$$NAqU7e8;@`
zTl{*bvT~<)asxH98e(Gr(h?5_42YVD?T*U80mLQ2<)yn4VBWb-*$234T!dhOi1-5K
zaUcUb7+R^g87v^7@ZKv%%_Bw^{j&Rf*5vT}t+ZTeT1%V_!_Gbx(uQC^Gf&&kn4@QT
z@g-B!&EHljG099RlJx0A4x*>U*^&`2fup0WO4dh5!1%^AKd_mmxsZq2Q3h-}VNC&l
zEP@YZ=p!xPk0l0xM+3}i<?}H&Cx+{G2TT38^Eq)6DmUCNAm2+k%3p?%iO#*;j7l<x
zbIbcElTpvk*lGO2UKMZrve!I$x9VA0$vU6W9D5L#%R~v~OnLEmKh?l;4+uYHsW|V{
zPJQ(U8QIR4+6H;OR_l)K6se&t6v%yNH;Pja?JIZZl1!?L)I!ZTiS;j<34<oShZ|Jy
z1gh(q<RQQSPoesZ=~98dIT3<WqDa~l;gF)WC@ht06}LbQE2uuD?Sq8@5BL{5hEsp$
z2ppdQus%BhEt;`A{~w#PODB2soP?e`WOd$U_d+?;gfiRS8|i>Hubk`HM<uNwY$^|r
zG_yx|;TC%&??|0m#6;0TGXa@etvU{$uj8C!e$C@UIoyUGr?o|&-*^Bw85Fn~7e!?v
ze#t}c^MfYWf!jOrwlF?*ZoOuDK`Zp<%^^s$bwC?6;@$%gTFR+Q@K=eqcDc7X9HA7(
z0yDCg4*>gt0{(&$b6y{6ERjM7rx_uVsoLJlxCFa*z-i*<%JNU3TmVLTOg1i*?eix1
ztR{`13JwmM7ODov6b=st5ZOs9-Zqu%>DX*AkEd7beh?~k+^zFqY|zK4ccXgAA<1Ek
zg|I*{#wE5O8)axAx2W}^XWUC{e4ebeE5H}{n6`NL;i(1SdpEhjn!1B=0g~Jfo=m{D
zl0dxlx|PNjPxUk12Jx-OJkm>Ophir|O8>ZBiWcGlYM*mrK)X})BYAk}+Ql!OR089%
z+r@M$;9=?<e+)z$Tk(+%%}zR&X8aw6!^b3Je=G0oJnDTSJ1+0NlxUP~GU~Rn(Z$-?
zRzL47Q6sHLnQAqPh@0E)RH966tj4bRY*y`OJ02~T1D{}}bLC36a9LInfu~q)R1p4r
z%HbNsDcXAWIb(BC?$|>YKagVw)eg%OFy&A@PWP!kfqDc0i<cZr=wng~6!v4$^KfEO
zR0f(@x<HizOn}qq+OFQbgiO;Komp5o1Qsrfh@qP0uJiGl<+yanO^5v=Ze5vWuB4v*
z|6~(`7ciB=sop-73wT;otK-g?OM@a~2cx>DB{(B{1?jmQxX>i#R6|J`$zEFVnCP!o
z1&l{@Gv=sxJPDQUPlG^#z!5+9DvHnV`Az8jOuq+^Ty{=nd&?iw-*^&Z<JScN@2=*m
z=x?^Rhg5f1iImG4$z(}Kzw42m09DRCLO}fF#To+rbqMqkxl^8yyl^auYLxJCV$+b_
ze`;}D6g<AA$d!{{0RvwipS6XBfpO)VW8{hHUS6cf5!A%qTDxC)Z<@SAJhy&2Ml{oR
zo)HQKcIDw}-jJVLt24ijt0`N5ya3nSn<;PV>3u<>5!GR0PPF9cSBoAzTLGoYL?(4H
z+wBJeh&LZ)Bkp22Iv-tic>Y=<;+ffv$;TWqfvfUcDVr^iHmR8fwYp;HHO9H`S;O{k
z!awDDdh+u-k8OcSC{&qJkCsNt-J2QZK`R9oTO4giDoQDCU>+eXv>;J$;qN09!66?p
z*iy!-hfU-=x+{tk!K?B-_XH2%zl&Bva<0BQt2t3JgNze#gz}>xigL{q1}D020Z%=y
z1#JYDI(}2s=8m0VjB@4R`56CiZM1+B?x(cZSKNC#Qu%iEPETv`>G_WP3F;dBivmz3
zFpAv5Yf>dF(j8S{qpasQO|3vlD8>P7XBRfpR1YiTG6#7G*(vYDwT_*(HIIyMLaWr8
z8M^N%4#4ty;Sd{@OZiFSz;U-=1Q^^NOb9YQWC^0XzY)Wm-jB?n68upVlJEX_Hf7WI
z&;&)9_1!herX2cG#!=Kvw3z+A=vY)Rx?sqh|J#`ieqcl0bUqEAsX5bQbJ|p-B?flk
z;8@;=1(XmARNLqcz^=#(pXMqu<X<%S9`#6T`&cLe44@uiN97gTj%coUUKJPUvK=%=
zkVaM3c3Pg*+{6z4>LWG2{m0fDvvm4rZz$E3n$n}~+mlvSeBMV9<1zU)&ibUQO8u~9
zmRfHH%<xQS<{HcqF+%*S(oLePuB$pArL~52^#pLSgFr*uy~tFTUsQ|QUaJ4}F8ylH
zMgW{A>#9#rhiEk_vwL>lEw*F*dg(<Z(^50JCv8*9{8lmGCUPND>X_kwY`FX2VcccT
zcfS|DKYV85y~(vq_3(XSWg3!(cW&<Q>PmRw_zQPLb%SnTlA%xNS*_K(TEl?ZRK7j3
zT~9Aj5Ot=ZNB^JwmAvPNn&G46k>sv)<!w)KJ3Z!?HpPV&sxc4GoD)}R_?etcr^Q%Y
zjyk*@5;X#tm;Tv|ZE-wj3RT*j*$r7no?7A44Cp3vOXyGKbdrS5BC~2vrw>q$F`?9q
zx0h-wvjBi`cV0ZL9KRqtD!JrbBGVC;^aUsnvfTJJHT?U_CtHb=6p0@_#v#$_I;LI)
zWa*!J+1K}z?<q`4ozb`>SPdMZgigrhM^9a+>=&$Z)n}IwW}>n$%!HqrrBG``-uQ1{
zqVAa_-z~cRPl{m7%T-A6>D3S7R^WEWm;#JfVy!k5j#A_<1iZwhf@)0__TC_zmXT+#
zbMIXaQsw8TZf}M+iVwWzUofm-RRC{WSCn(OWT>bQ2y*gKB2U@IvHUFw(BCa69|Qq<
zE-Yn-7gf7oRhy)=Kh&v5ElCq**Lp7ESv+}CX$<k1RPLnJWbcj09;&wql}>4$;$Y)c
zQr6f%oD>dRfVBa^cRvcf&ky{lQE0vi=U(dGyV7TV1hioYiYp1S-rNgxgz9SAtOcnf
z<O546&BxM)%x%R401WH|Phoj_ZGP*))gO!KW0)lm)qgqn0YWuakMGmj;)F|!|LoAX
zVy+5c(PYhPC(l+VO<hO1oitmDp~z@ayw-ymVDL$WR!5lI8I#;}a-W1O(U8lsG;Ah*
zVNLb7x_db@u)8T>{dF_CVa<gnvHm(^cee_Hv#Zj^VgexQ4U6;Sg`L4Jt#9-~Q|}H^
zt0=}kojK74_jfK1Ae^}CS3%rB)v^$yPn*)Y=;)E3D=hNF(0cxAeigo8d&lR(ovyy`
zSMD~DJNmk=AYT)^D$w`nzbfErkW&wu0y+2jALLUF+~p$}tZC|*U(qL>zdK~le<Muq
zbDW$~G;;OdjlDnu?#qL~O!mqP^D3u!_Z0O?IUe*LCDC+{RD=~53uSezQ@uuGE3kT4
zUXLB(lS*(`*ZeaYD1-NgBu_e^Fk_sXwpQLBBF@Ni(d5C)9u8l(^P4p&TlRILk1vs|
z<Wvm5#h*gbb@c2N{dB)IHfqga^yxz`{vVxO$xI72Xx)(XHGcG4)tD{*hLMYC%^+N*
zAy6Ftrcr_)KmEyt>vnvXoQys<on~KOJf7G0fL#3P+USGykfaz3)$ESJ!k#T;m(~j*
z<l-ILuZCPL<q==X<iZ;2V9(%_=s{bx((DJ?BSqh009f<!1M^y~5WHBZmc5SmI@qWy
zPf<ugv^W)OiX7S+tzJ|le)6{_uQw1<WU3FvgmpA|tmF@twd`orBRm4shS|bD`lsjd
zbC(6q{MLkK{@NYicDc33nG#kb6n3dc{r<Sq#ge@tL{7S`X^8~F&6DYK(1&v?2!$)U
zx>w2~v+|gA<ym^>XC^0f<`Jqt)`GE1&NgorRUwx&6J%}_XwiJVX(z%czJ`qZdU(zS
z&woYDKivDI^!eK!(qRn35BFXeaNl9(rS<Rm#9+!HWEvfO9a++-e`bZdC;Q{RF`TyU
z$rYvOd<K)pdH*D{W~F?t!-MNH<`MBqZ|f3zttBAto`bf+K(?}m`qEy~<q-l=v^}y&
zWB~6>Eclfp<-C`x9g~!{B5t9|dE7V9xjydcx+y>_rI$KItVfxxJ&0QXgvk`hVU_Rp
zN;aUXqL9!~0t(<QdtACrP`Z(MNC7amZ`nlXzr+@^swoN+sA5RbWMkm8DJ};L7@${D
ztY^BgHV#Vbg>Q_%0t)#8Ytw#zj0e*5oqV`X0wh$EOD_!1`Y3SostY5x@VAyw>+hYm
zZ<4tc*R4s5tL+$vZMO6%gqRTS=i8Rm2OkG3YhSBp2z-ri$#dcgJzKS{cJWKun^9OJ
zvoFTb!oF`20hsjFhtf_C#@^*h)%LA8``?>kDO8P7;f(@m$2_|g@6)s~yZut~aP}?%
z=Ha`y(eVhH*rBkX+dCx83$C6niOyzp+$Bm#c#dog#vzrxi3hh(z2z8r;)H)$_D@7t
zHeK@5$B2mVp!4D8kin+cRtibaD2N&cm7cS<Ihl&IK`3F`Mrzxvm)bj8paNoZqnIT^
zQ1z^LLb7>$)N2KjjeT}!d&_G!^9*rZs&imGIYOSn=?fl&7vZ(mXk9y|^Mc2BJ4BO*
zJ7ea*zo@6LeHt{P12DSB9-sWR@X+Vl?-#EWx$jRKU7<iJ8&KyUFKTyy*He8WOVat*
zrGefoz@_@g%<jIgD(!qOwOR1=hhp0wYXvn;FRETz32>aUzi=o_SE!(4{Rpcl8p5J`
za?q4os3b6OF)5O~ukxCG*FzJ3wjSB<9wGhiwKc|zfMZu2t?ok1e%mE!gGXkENLdL$
z@ooar>E^LlY*_o(a-oL3+vu0_l0TbYILpLuzFgH1J)NH;RyS?&YOTRXcAZAVjM>fZ
zNxJ`7rQ0hkQ0jZwVGu!7sCNYb;LGCeK8n-Ls}ks)o{c|`I(6J3J3B7h4Uy1VUt-7t
zV<$v*$xxHa^AAzVbCImsiBA>g+HG2x@MnmxR+l{+5dgIB7PF!~@E1~uS<!lljau;q
zz3uZaF!yR3=hURdjT2mM)(fua-0S`6S(zY>RmzZ-EmITxn(p+5Yd@R;rB*otK;mT!
zXn-j(xbuiTy)Y6Dv9Ps<Js27i2RuxA+O|H-6^;NgRodRL^*fv0qwQK>2a)h!Ko25>
zXKI>ylBck!z!<a>2I{DVb9cZ~#B|oT|H-rV{V)eM=EuegUp`?f_+`L8>fx6Q8W5Gd
zy#{9vvWi}^CKlDFg}_Bm+q>sJY<4prGI(Ly=rAM!jxab=36$!sl6|E*YK!*KXv?Gh
z(NGE9&g#Nh+}d7G>{eiq2)8EyX-}u+mE->THT<x)J3f?EY%Glr827j;*)!ban7clR
zTkLLksn6)dolFU35W;7?1Ug;Zli1|c6ANcP$>9r9D4{6PwdY<e)rA5Q43EJ%oT9H8
zlAhCCO0Kl`J^8VuN$}n9tvf+45TsIHkl$;X$&l<AJhj9RzNqqx;<*TsH40xF>hy67
z;{S#w!tT)Y)0^z|#K{EGwY-{D;e+1$ryq<PJzr#RGCq*mKx%7myzAj8P|xTli!Z~K
z@PA8-F<UO+>F;!n-4>SW&d!C%h(!tOu&7XrwT1y%XIMp#Y0vc_r;SIcdwpIgGwFEC
z4Qf10cr)ti;rF(v6m^nW9}W;PU&_y}THHW<Zo%Tu2=JOsQ#sp1U>6LZAB*|ybVSIS
z_q-bKaN*a_p-7;w1nWylFgP;`f1}wfC*?Cx#d@)b!?emg(ne;<{rfslQpqpE<vADF
z2@uBYWn=#;TN-VRz`8ad<1gMNUcia%ySf~OTl$d_3x2tssB$HZ?8?TTEL%b^x%Pfc
zJ9`f|eR5Ij;hFf`l&wlG|8~%KB*R@AA*(f}Qp$-HeKOn&FkJ~;{|vw6nr_i^AAQgy
zTO+js{hir+gyd<bWbz)`_#CbVDMNs^@O54L#v(!;6*poSFN7Hm?qBl)I5D1Y*-RgE
z^}D<+6dga49{}aaoyzl0JVct2fYnV`cYkLtxU@`a5<NM-Q&=4!#SL_bmdmGpuQ3pe
zMn2icl>1_kn7`75G|Zt75Iy15@%%Ek@{3+dR2W-hV|(Pc#TCTW?X7}$I;t6*pA$^h
z`-0mUbORdVJ(8TIsw`{e3otkD2s@Jszi$IwlYJLYdnM++zzCsL_0LY_bw?*4w_fhB
zzoY3PZ~Pc>sDiPH!@BmU)~E_pj|J1lyUL8Z+T&lfk}qNjv1P0*fU}$5N4p1{^E_qh
zt9`F(1peUN$GwZ+ygofdek{g`-E*4$)#Cq!J?F&~e6e361RTo$4dpr{D#Le;+En0*
z%;QS_B(rFx0xYUsJcmxlBH3{)z>c4__1zNhDW!ea3$t_4b||3#6)rq@A-}x+!t=w6
ztu$4)nPqDM0O)aXkncaulG^`Qvt;aRx2YtvI$V0Ka0BjB&lBauC62o^%J|3Nl!Ihq
zF!F87HD@n8a%--H25F-uBKRp0?fF>w{X>D^8xbV|#pA-v1EwnvoX&4Wm4iF4L`=w`
zABrWETYp1iwi7~sQVIMDK+CJ@&hc0iH9R$N`n%#+s}`JwuA>X*>FhM>txI~#6`QbL
zFX%<(;-qYE&5)Zj-%Bot2YRb`EQHN&#ugd@@XA@T2QUTH>6eWjO<*;abC<7ZVbP~Y
ztvSVY_@u?+*Z271&o(Ueaw{(W6jJE9tf|oA_)GC*TzR7gDM2ds3}77?cPDxNO6!TF
z+iqd{<}K}^z|DUW@6u5XupC84#yH(C0@X#fxk96@CC}E)$2>blqHiqTjQy6c+mzdH
zXT28tms=}qv#G-$AEWBat?%q;2_}OK_$r52wVYoa)rnev_n_wN=3of(8!0a<6{O-X
zMy&n`{fwO*Hi!G|O|tW(K~xVV7|2oYUZ)L%`&f389BVw-o(F%c;rznOle9%Q%l$ee
zewG~jd)$MN;#^77Ta@c_VB=AV_SOtZL2FQJ#vU9H@?#EoAp>Kt-rplo)Z8woUEITO
z`8@hfaHzU=>)hpP-k^J0Xs-Py21nw_v?3`$>9+XO{VqX01<dTFmgu*bqK%Zx!Yj~>
zYekQc?5ZT7tpG)Q_^2j0#%Y}3;n(!@Gtlz5@y`#pfFEu~|L2*T#RU)*_l3H9VE>Qo
z0cv3`VJadwlyBCv=Yn#SoTDs$Ue6tD3<Zs!D`%&zXR_uR`6z%^Lw2QwI?3$Kyjt*P
z)c9x_(7yF5%RnP}9U=KpR3P8T6Zr3jBKx&i4!05iC^L?9=MCWXr#2Uoj;Mc!mqGc8
zIsHLK*L?LkgxN<)@zo0tFYjK@tGA=gl@rex?PmA)#7vTYRm)*GCpEap-q~TCV($Rc
zmj49Lb7@92|75_-|0_*M{x9i<{F`?D@4I@%<GkgCn}_1$9&HI-sN!bd#(ZXE2eucg
ztXRD>7211}mP?Y`4QJXbh2^?++=l_i!sMw<6X#^%RIj%{5vpo#wG;4DIT+%n&W&vV
zWI_)Gz<FTKWp%ApqSID+qf4##pLR|(Kh+1Nw_B4}ky`t6=Ai`&{#<zXDUlI>MkP5c
zwQ7kv1PooMt}2q%3%ZVp$A&M|{?7kk@65xYZvS|n$Wk$eO18mRi>zfGgb}h!WzP~4
zk$vp4?~=8wX`!-&>{}RnmQwbev1H%I&bhyyr|0+l&UMbY&Y$OxGuI_0W#*pm_r5>N
z>;3+KAu}W$_QGM%Ae4$j{7@Ds!`h0kVX0@`KbQf?vLkBE0!826WPr7p1;9X(k69X+
z=CbSQ0ZmGo-3KIqbs_{;6dhI&J22wJIZbJwjx1Hw9ei~zUs$p~WGWaiDTTU8n>$m{
zT^n-v_9#|i2`I?+JbEXA@%gE#(1BOp4*?JW9RvY<>w82Y*Egj2_ukuc*_b2g1g>cy
z7f}%b1e()AZmAkn1BY-Mbx_7xnLW1)p7_e$g6}uk9v(qmi@ch>-}5Cyw#g7OaYEWn
ziF)#(Sbh@uW_&Q2c3dj%>kP&OXxkrZM50J<ys^RFo1B8su|u%VFQ-}*%H#A00oJv{
z6K2bXy)T{ZtrP$K;rny-I>h}s1+z^-Yu29Tl?CT1U!-`9JTfb?akV>0x>a@0VB_;p
zBYo+JXP9{%ITtwL8b*9gYgOybbd5_nCMY)0_~fbEAf#bT@iCM*uONU|W`*d7zhC&b
zp&+5@TZQfSIf)aQNHdWFu`SuFV33FY$*`9ms35tEOcULMDu$PC-93L*gqUk3rKn5F
zBT6U=5J0Z@9|l#AkP&UOP!V$@J{ons6Ds`Z&Q-0{?3V-`-9ZrDHC^Zs8AGL?>I78a
zJk&SfY?h)9o&;k;{(gMwvJK5J)7=Nxi?oNR5txnD1I7`FpTK|++B6RhgGr>Mq+<e?
ztL_d1C7+iopy1S3+`2qwM5k7z3to_lCj`Q;ho&qyRfD-e9yb=Dbjs~R5-Q|%LQQ+{
zEc?XWfhYhF&ikF?O=^b*&#6Q7<03&DYn<x`cFX?;zj{AV%7(tE5**vgbM5e(rnO8F
z;4_<bR<fD?Ze>Ma^1_wlK;7v=-!`~$))_1rp!PbC0hLT6i<Ta~lTf=Kv;=)FspA?Q
zH*<<QU$rA6pqYi!aj<C3`VZBj@yed*_Ovu<Pd}>sSpj%K#bP1pW4H1S7dM|1hpTFy
zCtt0{ZW^UU6}U1#dyyfAfBV^HR4(eEi`5b1DIT!3OR7v%Jj<!sV@afCl!+&MN+}QB
z_PQF3FsjRG85HJ}qyDi0G3m!-rR(t}n1Ah2pE|&D3tua~Fhbo$mO-R!`i^~FWW=eb
zYL<iLM@>2kMfRN8X^=1O)RkeV-8Rp3G{H2}-~p3DT2OHW;tH;yzP|{bTcx^7ocYO#
zEE9WId9ylfirMe%ipNAFYmi=zU6JuT)kA7vMF02WUh;?@MeQY+)VdP}CcC#5?SnFe
z3I`!pwBpgKnNz-foX?ziOiIVH4#gyQ=Agp(!8mdZM(|S$V+EHaOK!Mxp-Rn5t>uEz
z;G}4f#?^1gA5TN%uE#@T-~W2#@PeK+EZY|$>jW||)zV>y_C=hMFPrxu-%qI&1tzae
zo?BF+kO|c{qc;ifBuL**$tX|i6&^i<c^c!9TLHT#<lMDu)!VyBM93yY%#9NMM7&<7
zR?x*{pP09RyxkRK(OK;Hn&V~HS_!_?(znNuSyi>Xh~J;x%#x)Kk&S1?$<u?lo3*Zk
zbP9VJx7!gbZdx=olaQ-+kQZ_H9KCZ?bboAFOuOws{x9x4GnAsBDeX9=07CTfuU}>O
zXmHc#I<7!{@h%A{FYI4@?vP`9!){~rwoV@`4>-L@I0=?4{1V#+Jn9g)`v3wiW4MDk
zTg4O|jBtMhO+m9Dc>Y}J5o&+2XjTbWu|vw0BAWLk7vU%}o+{DMK&Xc~6HM<`fdZEh
zS_6^#cMj`*lbTN(nH{=f0hhs<BhC2kO6P#goyNbRa*!?P)L^s~4`D<5Se?b;bC=Ob
z^RKx6{@&oF3cNiUN}L9GC0sh4mkkZ7FBZJi^*CA@WJw4$sl41rk?LbctMii~L5(Or
zz|X^(q{BNMtA-p0ZAXdlL?ZRHG_;oV)vI%Af!mX7M2BFgy3o0IX>;Dth{(yA`1?j;
zKF5D$oO2MVI4~4P`VkJzzk6~P9;e6ylYJzs#d9t^E67{X>}89K*-Kh{@={Y_4ETs8
zZ!UZRJ`38o-U-j}bG%>tu$`H8S**#IH3));Pf)Fs=}3!FhcrHGG#iDOF*q`i-J6O?
zp$ZJnvb@B(bj=-brYc|r(!OP1KUD5<^ae?3mId5N6FqHzn<#$l42=Q*NugR^VgJSL
z<8WX)#^V<yx_&ZcYF5s^X8MKldl<+>k4?LNNVaBjW~)81(%?DQRK8aKsdGmlZBH}O
z+;?d&*PiF77c6FYpHd!8g++*Y-PY+LX9|nb&T4o&(LS`IG4@A|Nu~1OdtNoXza(=g
zN{J^;A*Rh&9Zd@l#I4P<C}0&$c+5F9Y+*1Zq}t&9IMUmz(|7kyvP)dSypSJ`(hTF}
zy=46$S@IfS7PrA{L6j2iQ%aS>5;Zv3gF7c~@Wbxbf524-+(Tpmv%C&Unb0!vpRw!P
zx3a*ovq<*X<6McgXW7f|01-N`Lp3eZaHQB-idq$E8<I9vpR8z1xVe}Zf^{*Z1Vthh
zb_2e$Bg^K_@;rK~H8Wj)h3WArX6Ax15KnkQNC(aA1HBNpA<7qeUUcn#)+8m$9{oQ7
zL`2p;vh=I9AtsAm;QJ4}eSw*o)4s>af$MWq3S7ZzoGDN1oxoP!d3xqUR+_jp+oKq0
zhT7k(ER*t_o3-8`Z<?p43pCZte=zzdbm)47QCY>+Nxi{p*MBhIuBR5mEoG{dqw0*x
zQDv>$0L!Cty(Hfa)$hIayupP<*d)Bqs4phU^5kcSUx;Mr82PNb?2C=u>JQW}K=hF+
zi!MCdv=vv~k;@ED*QDEk&8y3rAzSiiYA3FAUUsG^NzdHiW4R76#2qKXUsW9*F74ix
zWEcNp*s5*p!ZIp7!+k?q>0{@VxC#<ECw}<HsGvfl0~FOGTxIMiKZY<}N=r^?rREDw
zyUG*wc{1q>m7alUB&Qa&B>5!7g&Q5&&R5E%QMh5&#baW0So4}3WjJ7`KYC8H)U4l*
z=6~!kOop87q8;r8V$kimVfuDO7y3`Pgc;ja@bm><0P1xP?)zbDB#dq>IG%+jlbuVc
zx^_)@TQXlg?L7L<T3O}UcK@EEl&ahLP^?1(hy~B>NZ;Lr(kgQb_3mkbtB8pWC=8gw
z#KWH5K_qmA%Sm#Cgvn8|=bV*@vd*t|xpe0pbZ~doTS1zczaH4fj$g;@OcwT2o7OLl
zrY5|LeiuxnP@hG@p!>1&;yxI{mdvDsPGY;QXJlL}%sYg-6?ie2zx7S*=c`rbO9|i4
zNhkbx8go%WXAP5Z)=XWCWKG*G<kcjl^|cO`v~|+EQn*RO-z=(@>0WrAPTT2Js<gs3
ztI#vg?s;D$4>R}ea8MU9ij<oY83KEmU8rMzwfRZ5+kO>k#C`AcNw(V*mQ=zLYF_90
z)!84H#BibC&f`a8;aA&^8;mnA784m2Y+MW^M2{<6R!5x)_dDpXClKiQTayg>@4j*y
z8hxrYY!3hFtHm#U9<4Z;Rwqo%@OlLbgE~QL-px+EeZWgik=6uzN9NcAw`NtVV#)#Q
zV7UYg{7oC_cUP(6mvt>uZ_Jum?I%r#O;w|?nrgw<LvuiR`C8!p_mN-@@(F|pQFA9$
zA{fUAxnAm=U~~ibk#oG#m#vsz)mIpctX6Fk3I-_0uTII_r1kIKX01A#oKr9VLbN`-
zl&geiW@;5dXG!|Wh=%VAHs&ZqXkC{lahu-Dz855&x+X>VAI@41g1_q&3wU<CK!#fo
zQFgDuoO0he(hL`toCR<9y=w$zsTW-;cZ}Ko3cEMsi($7+YexdWTK9gw0J}v;DSW8N
zmGi8LKGv0^pzCdH>HD)hqk;UFM_%%a-qp|v#;x<>`!<C2zbRC(M(z9%Xf>2*X^bfq
z<e4eb$x}Q3Pso%H(>nc6gkk#bUkpY_{K-hlKf%W<$O825hmO^M*aGDL|3`l>(f@HF
z4v)$Lb63n@8!fESDv%grbJ(#<l@sJ-PJoP8g>uCcV$TK-0n%x1GcV^qH&7&lVUuO^
z-Ey#zy(ici9F)f)yf(kjhjOMRb(I1_Z*db;KJB5sKM|U;x(6UI{-KqVG}ZB&0F+r)
zLzH009d*eY4@9a!x3~w&SWi+(NIV7cg|_`mX3>(u5IEi=L&A7{2nv;Qs>*P$CR#1i
z(-9uO2j-80nZQ`motXJ}=_7l0@c*jXNllc^Cy5}W7@hlBdRR!71~xuhU!2_sDeP0|
z?0cV;?(KeL13<6z6N;~5AM_XYiWU+r(?_&A_gB(keRFJFee-$F0o%WQ9J_!g@b(7F
zFn*QZ5u}!?(BM<2dNR&$X7^K|jTzDP#zsmSXqRUAXTHpo<Q18Yy{R=k$vZc@z{t(2
zt)WO(xD(+YfEXf*Pwrjl3((DcMt0VbuS!&M72r|1+b58YSx+hqB`CStKh?vXe*j)q
zj9rSF#C+NO=2Y}c(B0~vzt)Vn!RXQ=(?hUdi$C-A;hU2r9}*Ek9MOqjY<Kt=n(Nwk
z4c!g(rcU5|D0@nV#9`ki=OD20MCS49{sT#nAE<7K5*CD?+5?I)Gx6htjoIy*eMr8-
zv<;MJDYNcnquY#LI6@#h+01jTw(eX=(6@aF1<swa*@fjL>-y8~zoC4n0TPxKanbFk
z0k{pZx~<>8Cxz!<#_Y&!HGI57Uc*ajMr&N8cH8@>!Fej#q7fZpgGJHpigDBbB)#8?
zKowGUbozBNAm5XoSJ+}8zLQ?VY@Hrp^!l++Nd$rqXUeFOV`hy!tu9H|9xmtuC(wbd
zo_LXAfsv<O{s427lp5LSj|IOuHmO(xv=#&ax*yEXzHfj@V#H@wui>jmP_l+uKxf0)
z1JN70HcqLk8mQgrSWAL9!uk={G0E`gqb#pQa90$aa)&xUlW)h)*5e@2+=C)&@sR`e
z+YSb+l_m)B0A;IuNo+i?y7q?le7V(Y8wiQy`BH)BY@xarIS=W^I+anoW#1Y05zyh`
zmDhz&NVG-H5ra9<?YEJj+!zX2Y&NLU2)>{KzCfrhxn1wnBMDO*tHzhi>-P-{8v1Pe
z_YnGE9NZn+XR+TcOBvK9kL!87rS_(adfuo@3Q5-DiW(>+ABwM&>vy)3Fnzi4&aip-
zd7eGcpQu8-50Asin8s+KE|jDDM)DXcn%0j49*36+*p#>#cL3&!ig$5CO&E2igK-ke
zi5Haft_@Uf8yzfa6)xLGuh6}X{-nPL6?HN1_7){UhVI$XC9wsKac*y---6_9?M+2P
zTk)Q8GUYt4qrJLnsAMX||H!WTAC@UShIQ{F4jVS#Brm5xoLpA&vfHCe@(;Xm(Hnx8
z{Hf}x0UDzeTOYt?hS7VA7#;zXZfM!(;U)CDVu>=%aSIyF*bq@L2=5yh7rg1VyANu`
zA}f_wq6<{KLsGDs%Bj~pq_>A1v!F7D5iuWvpl98A0`quqHkKsIvloHZ=$8uVzOBOk
zD9n;u)-<aNBGbgu!ED#5Uxn~{%_Va6OSbyh_5X}#E@=G>%MNB9@f>UDf!P6^6k`8o
zAGo_6dFt;h2ZG7U5I#HKcFjJodgExmAS*r9$Qu`*%i=;V%MOPscBw~@2jAYDg0y%w
zBoV28?`FR~xpAA#d3A=>kBUPQ9UAhOwo;CP1g4tF@jdp@?G!eZFo_=qC{N<{amF8X
zfGWvA>j7)71Wd%b8hb)p@XhuqjhBm(kGFKHr2ca@_CH(&n(9`|_hXW)bDR6kX-R3`
z07;C-eq+Q4?)3`p1|O~dJUI1kpUK0)eGNB4Ekl+})L5XB+M?Lc#sx}XsrkIIoN81e
zp|JsX(lSbRS+sa<g(iGVv)(PBZ|iI7jPiCoVTq7bz;QD0sZYLlikp>V+Uwp7<nuIf
z2U@LNK7<WN4re2^9*cT=-mfc0>5DF1cRFc5l3OfQ8^0M!KU}ojxKo=~d$c#7SGoL3
zq}forpEg70>#yA)dfTLiBZxBgK3sUvrA{jp!nG5A{!M(e7s4Bwr5Y?V6Z%lw)0OrT
zt@$0|iD{Oooi_%{r~d?+Hn(`}d=@`gX6+-qOx^K;i5B)7IB+I-k05;XmY-ZX<SwGz
z`#Pnr#<%pIwBiz7<U1UG{{X600LpGG`39q#9%6;r?}v9tBJ;hxi-^F9YNcA;PlUOF
zAxc7%y0;${qtbTBWIuA&LKVVvY0ffPF|vX$Y-^+$lkTcSUWbMHl3iuA)a+?Ht#c3`
zq5fchuy1Axh*I%YySbcBWX(G_b2cCH5>7+KD+WCGgWWcUZHUF)goLXI6;{+@pK&y)
z-#oNk&wm0by(gp~O(GcvT=~(*kI^P`*v|BWY0e&~ZX*VB3suj)YishS3M0&#f3Ozs
zCWf-p@bF(Q-P%YsslDbiDSh<KvBykZx^$?0qEFWGbzQfC=+`wO+?n53RuWP@7+=LS
zc#Nh;$v0WnB);$4Vt(O%naV3v3jMH{!5fo6b^V$Tr!CdpL%wULdDddLxthh(-k`gQ
zy8A>bhxZ$jvgZL1k#FIjwK_DHpj>#Lu-=&2fbO!~)9K<?q1dNwmnE;IvJ8&Nlv(GM
zTL^bgNM6l{#H|#3X->42&&;2fQ8QC?{UCsKL3ud-$jP$)ShyWmkvl9Ednv!?PgvJM
zYWN(}`o^)B)K&h>NIkdyWYy++`$j<xwrpYb_uF;@Fu|_};+f_lvyFr2Y_jZLK{K+f
znF}`R!^5=L(g$<7rK949pvRwngAFFWUU5&Gu_fAazB^E~?)!6|cxh0cE)5Fn<E~`)
zqxh9BU~gifBHW7eUR=G-ThH-2?zPqk+3V3eE94;{UU9CN%M71cofI;4g^?y31;#|9
zuqFqVAT)AnWl#L(=2n<g&G6Pr7gSyh3DVvMXa@^o==w<jq#-s6asnZkj`{YX-oIYm
zsn;vr{;*e3n1(uVp09{3g}t~kcceeb<kC_HgcS%>?HKjy+vz2xbd~@kmjUyl5;H!v
z!HD=;BFzbmwXJf_FfHi^hPfQNnemxtDtLx>XL3orf{)Pm3(xAfiT(y+?3Ls#W?U3_
z&u}C*x)OvBr_oK@pxszR*G2tvL8CvA_2K&0FR(20r2qw^mn)cNiJ3A8p5}Q(caxZ8
zB5w5xqH*jvLdP9@*Pdb2(LW)0^~a^)*=$iFO~Pql{J(FrIN1_>KYJ4^kQCHs^(XbS
zc(&^H(<E4plVPiA{y7j`$mh37+lJO2Z@y=RBFM?r`L81e#A-sXi@Vhs7=}qxQaTrF
zTiajE$ik=+8ff8h4Jj-0C~Rvn#iJt{BfXSgy<zWave>`7GbB_xzAvJg!{{9#>$&N(
zM7jbB#9xwnKs>veebxQGaGHER7q<)iD*Id-d&ld1#}0erB+{bC(rI}47|WCh%;oa?
z$ZP^R6fn$VMr%)~i_gE+?f&vQ_i2C%*tE47+^G1TIB-Wa+-ZFawz5Sxzjefoeyq@V
zK*6xlz_j^S(er{6SbB5FWdeHi^H2&zOHO|NPhjQ^-bkMk)<;}Xaunjm;=6OK?_;CI
zOvP7yp<Ooys$d>eT_lOIXNC=)iwqK-m#m9Ro09NwqOzU8Y0cOzn)m%FX{-~=1Wm!)
z5h3%Ck66_TxPpjP9yvFR!U3z*6;$BNsv4Gc706jDjH)n)1z=#as^y16v{GJU#LLx|
z@$(d663@OfGh?V&X!!|AWE{>@&I`Ev2Vt&Wlg(Zw4kX}=5H(07OL}wuNJJ-q-acTv
z_IkAV?`$#EiAHJ#aic&X{(p+RrbPwENBgP@Ox|;wB_d4H)eB!@v%u?dXwUK){*WsZ
zEp<TNBcU!~C4QXCslv(Nyki@lah+O^qPp2Ty@dO1EzjdCXf;`OmKT@gTwh=O?nN56
z$ugO}P4SAn`8s#ty!;pXR~^Xf^lrNI+>)Py$zEGGM6SC%Dj{+5z-3!5$bWhkZ}vzX
zzW+qvYhRT#Z$q(WF3APB+|qpO4(*3(n#*4#Y4ge^!*=%}(zxfrLF0pibF5#imLO<L
zEEOKDke$^;EdqEq^)3$&-X@dbv>o0DO{x<##~(lCy}El;d$drSmN*hEO3d->o)j^o
z-M#($m!0SdR<zBpW1bLv6}qv#phFxXCT(MrSXwM#qattPk}^Ax7yj9vphR<a2NaFY
zVPXk`!#zaYM_M9vh*hLINJND`>(7oB#{$ISY$fCDgETWxB^8p%VHrR{!TyOBVt-H{
zC~>^UHg|-&x?sxn&lQ6T&BzYn;I6Me>}Xa^ciR^0F}IWU5<d(un~{`pRUl<JJq5u!
zx_t%jUgL20MqaU21ihZ6WVWkXYjV4|0?{$LA)CgTP&ISI6PWuKX%e5#UO^EhtI4t!
z=1wL#R=nr*K7rOCM0tg94`O{sXS!F~vx7_QM^<zNrs_tT3ZBr7LHcy_aUv3XPs^xr
zXV*a_p5eM!AH^3EwHWO0>eph6aQIKgXWaQ!4#JptW9@{}A?}}PDd!^uyrDC-nh(-J
zcOmDH=Tk&mV}A;|`robyxU&Yy*E_&hh0EFf1Tni+kUs+n>9!@!1Bh6&ggY5C?;1%F
zuT#pD7Jpc_HV?^IUzhtmp}u?*ZR2wAyqtOSw?flp8$kepq+g$PFIlK>5`zjnzDwdO
z;!W}ute|EGs?_Gq4KB$BO@-Zn;CV(<+CST%4(JufKK~w^$e5&|M^^)MF4p3br4t<~
zq^C3Ht76Bl8@enfXDN50+{AHntTTsbSDIPZv=oN6#p2e9$fsfm4Xk|->EoAG1#<qc
zXLoExKiC^gH7cbzdD#xh`CVm+&=E_>i7uFZPvoU!K=&fpj9MvI(Fm>kR#l}MKWIIB
zJg>CLRlhK~60<FrXkYZkzQ#_s%0)*Uz6m*RY+H>J+}>e*L`NEtBA`#zwG+Iv6RPfM
znuXRN3|&{(hD2Dka`2xS{QTq>Z2OYsiJUEE5oTgOe|Fr?^&jo|D$!Dh!%how(dr_j
z+jCAUX6mZZPj-)A(ipuuD#L-Jh9V_OYVrB}+cy;Hos#oO^HtZH*Ay6p86C#0srMCc
zeYxP?1-1078B)fAzRr1&JX?eEy1-7BM|!X27vJ{i%yXkuv8p8E(Uv8cx#HYQWX5uz
zzt8}yie25i`vuWnHh6wUqf|n@bL8{Q;1a0PPXF4rADhjQ9)rNLkFGyt+-;mOo=0n(
z<!wx5p+A4Y<0{Jjr=asBo2#X5x0KmeFt7Z|;RUa>jPu(-RklbW-+f}-F#$Hts1p(`
zk~4cfqM8fzNty~P9gGe!x>RGSg#f=p=xR-fv<AY4ma6F6ecJ^<9d-iwIebIsl7Sgu
zRX~NhvpSr*E2QH*RJSd2dAiKriMA0Y70Gi>_daTzuj5QlMYvDCoPXAwo`UdNpC5>d
zMlF)}gf=sXr9PcD0bF^FiTLmCYmQ~Qqz88n`sy!pUZ?1MeEst1E3+EKQ@D4n5^>bS
zl#20*x6qaFE|1HWSinV_Z=pXoj*;(n$CGCLc(>54-FsBht+xyW%=4wShf4+=W6@Q8
zL0dyu&FE3I#*>s4{Yy83Rj>J-T}B$3&|YlN>^3px^$U8ePWrBXXsix)Rwfp{e++SR
z2LoSynbp=WJ8%1vK0em)@M{Zy^w?~uQIzNwGW#GTcR1S%%r+;$pU8uZY1&I%)HTcp
z9>L@8=SZ-mJ+v4H8(Du2JIq^qSU@NW%Ek|*A|v|_g>{#I)EEI7`YES3{nKBIqu32f
z@KEuaeib{h`Za3k^wk7%NA&1p3YUcgWt&h?mnp<mwR3Oo(mZ%|^cgoe5u?DA+}H&4
zm=5C0K;!>z1>yHaNIhmuH)VlQt_&GMCn5EwloDa+kt~sSk)~0g_JTS?Suebkp=Uua
zn7(gwCbND&&7w#+qAHj$02ploLqNf+3z;PsJea&@2A>()Jch>++!_3^9uX9_taIIo
zx*Ku>xx`BQ5w{Z2d#mE-E+h%to92`jqaWi(wa?eCMcREc<%%Ek&O~e2A21IPC34P9
zrkX!KmOkE)?usRDUoO|IZ1{R{(D-vS*Gl*I=fvIlK8FjsqIZqxH}qU;!NkF7+mK4+
zUt)d{o~C;gYKl$Pz?(q+`4~kleP`)1|Gs8q|6*AU(zozIpIXTl3ma(T*rLj9pXRDo
zxo%vC>OC&jNLvc)sFuGv1RV{#iOZ{;@XhQ2Z|ZGR+Kd}zE*X-Gs&J=71(J@RO<&N;
z%3V*qmFsTI5s%)PILx^0%R56BITiJa>E#-c+x&;BXV$0oCj>C6MxlIka_$^wC}N3@
zsioer@Y>&EAfH(-!u&{hANm>9G%d?|=as@>D)XO=0eLUk@ZKpm6DoQ}qFY9<@f&O!
zQRW5#O)L&noMcf(OE~5<+=6g3yVlv2E2|JP!SiU{oiZw^^<2G&EIUlAo}1to%Y)<g
zh7pR71)f#2yK<Y_ylL;+uJBD9=2O%w`s6A*|4Gl#IhR0et%oq)&|6uWuo|d5l$QM2
zl9mjD_O6+2tg`wT(nfqS^vW{CmmIDq)}}PLYs8eL?a7yI#^nLd#fw5AdsEszHpA*l
z<toJrP6m5nG^he45&C|erG)C<yC?HCRalmQ_1hN}82?oSzL!jR<PF5kE7t?$g3V_V
z5;kTw*<6*U+#ao;Ek+-vICj3=-eyyYI+P|$H6zqUyg><tW)j`?uo&!%4YAU#st)kv
zrQp+?Uq@b4Ffi!^&I7xxt7+>u+HS1Vg0S#a*qaoT${4<+w1Xt(?jC((cGeRD(l1Y_
znA%S#-|)uh(OcT09|Eu=V(fCQs!hUD#g$711GIhGi+#=!^d(D;4DEi0Nroc*g>4D<
zpk4zH;bL!^6NRG^e#nsloMTJ;h_lnx3&i-!7&--PnqTiY&<`8eSd7jvRDFk_f}2;8
z2!!!tn=fT%H87KzoK@7EZhlexX0qbE5dquf)v{z60ZM7<X5->Pr-ZyiurU-eYmgvs
z3e+*L<a_f)ZMSjcG2MtYO}|3(U7JDOdTcJ#{$KT|!pS|TO8LhKLDzoiIVe!Ifx3?C
z?))mD^rf>)ej~7;?H7XMDyZf&&)Xto<~%N57`@48Z@?D0YDB}m>@g9*v}-(VlHp-k
z)lFwkZ$DjLVi1rTStB|XIM_2`Mr%@3l?z1A5$@uiJm(}Cr4lZluDQq!mPkWAINCfj
ztD#x2!B#A9(`iRXcn)#$4S#oF2H$&$9~hG+dNcF5&aPR}R^InZKj%8eHyXG)3Om=;
z)hLP*Q`PRf<K6h3lK6{LpsUW2t>4XV1Bn)efjF76t+rpGEPrl?$t{`N#{aoq|G$U8
zUu%)`|N27w<yNpl#Rn|NATKf%7_hwBTx|e?RtS^>MQ}9|hNoXo8WIqZn*}v~#UZyb
zpu~O}wE@`>>_MVsNr)DJYzur!T0rF<l?wc@dR|(l?LJO{dKm7&DJBc5<W_)GDC?w)
zZ~!_WsZo%#0i;oglG>Y_e=bMyQu;olecTdcPqN`Y^U`w0;RMYCb!L%fvGo?#6XSxY
zB25(deGjPZNCDB8qoaO`%b-<$f{>|DgGHpZnPmjM883iJN^zi9lIla4Vt1&vqZ$Sd
z7Z>|fthZqiP?-vBY^h=ejS=Gz9d`qtJi7V*em)M+aEo|g0+x&b!YeoL7ViOjfZ74Z
z;$>H@6Uhpc;thBM7VAN2|G^(p>=DKFy2u&d+A-i|ClN_t;paB=q*N6s@u9|mh^i~7
zAM=PGLG_a}%477NXbowOQEC+*FqlHM&%m*RVf0vT*lA~C&3^;p3e!$eUMGLN3keK*
zpjPt;P**!5>h}w%Zuv(l<rON;LcN2jVb6_@9`M>X9wXiuoWOC*Hf)f&i@`1FoHk3r
zo--yuMA56%8c9CYrvsD}w9IT`+ftx6QPfd5QoUvdISENN0LW$`PNZsr98hxa{#TTo
zqsG4}Im}AMv%_Uj)7O_n+Zg=^Tkfo*@8ErqdfDJ?d&4>e<?~#pbAk{sCr}qvg{Zli
z6+6gh58{`294PYDWACiojj(*Xy5infX(S2xd+ly$+Pk$Xh!bf<Izq-}X_>^LEO(Ev
z_e~TANZ(<5_3f&iZNHU*HVlAu!UfXdCS;%M{|KAV*vCcsNUjx?(f9x~A|3+mvi?XZ
zvj7|J)dNY3bl<Y6o#j06{R#q2E~T9+P>V8>Jn(6|lJ(7$y)IUPQzn*#^KO!&JA~G_
zky*D&sw*pZx&>>+9Qrgfx_+Dd@a|z-rop7))4DT7SyxPY3qA8n0|rj27|<^|hL8c6
z@L*AxMk`^QXscY0po@ol8l)y8<4ocAFsGt(3#->6OxHCsuCC=rNjJ;<Wc%namQaEs
z{ZOS4`UWKqu)@;L;BCuP=iU@vf}bX}8oO|@_&cKJnt4$9ENbd5Z2;#a2bdQ3_ajHR
zvRD`PET;6mByVU2R%P;CHQ|y>-O!J^h1gTKodWKk{<u>=xF437-7s6j8E3jiK!Dan
zeyGAEiBDV2S0A86LrJo|3^37As=0xjB~g;(l9ZuIkJCl1hM1hD!64_qAD>!*^B7iL
zoxua7rxIYGWBMz|g6$x@!!<M!t7`Pp%)h5mQp!u=^B2oG=n|N(JuE#IW*Qy(<cJzj
z(LrR?xA*u5lID)Khku^;aidlzdu`OX229Hq({`>@$?3jjje6R~mHzL=)Foo)pkiv)
zavV(vSC5?eZbc>m1BDQE71?QD#kM;=FP!tS%F6vdz+y@bs-@15FvDd-)1%hj3b}~}
z^`G_2Muogbt6FUkUxd}e2wcW}($6WsRsD}8fzA{B{uDxsAeeGKbZM$}k*GV`Ytd{&
z>UDsI?CeQoi{ym=a&SqfmTfFj$TD9+<UK5xu=~!PS7v(Y+P|#i@SLdOIfx0Zs-1<Z
z*1IN3lLpSe?)GRZFy8Qj#+xxkTxWLQ`^!r%&;4=_<fhGs9+V7qcYj1}a!)^-&koc&
zq6$zYNuLh`f+HZJRkUkp87s_f(_3<_>KJTId2bA010S1JjfT$Zey0KnMxM*a*H+)6
zfQ8lKC{lRxMVE!)1qPeZ){|{3IIe*#((&c?z3K|@r;KnAI(@=c7aG1A80}`jCy{;}
zhnka{?GKI|@H3kY_E7y#JxYX9Rb>T`CG!xjQY{AARk}s1oep0ieRbm6M!*xrS-!sr
zu;=t@N$KO=t`!X;FVMXZJE0ayZ+{u!0m3Jzk57mym?O-;J{-o_Yl@=s_5}b#qW<cd
zX3$bzDU#W=XtJu^+mKUi#iFu+r_9#4(kRt+zz!=|=bh>=aokim@hb7qZ^Q+tiA>f8
zO+dPf9D{m_hG17LBH_~f9@|g$JZ;C8`D2QNO+ss>s5rjgZ1(A>Kj$2j6YBP(pe`1a
z!x#zrfc|IaKbovVCA*s48E|3Md(SIB7`y;X&4kv+zki24<{?~2y>zKGGo(KU1fAb<
zipPM;BM^L~+K48^gU|8uLxyy4ZS8h*H{T-&O|~t^CAf9#CaA?t&nr!DzWltbC3>KK
ze)^pd@jTd?g~DUeKS6KI%pF)VN>?@=fgaYYGwqPuTM=mm7xL^r@ECI(y%X2p>L16y
z2c+*9ifyPHgr{Ic*q}9V2N3F4%lXCo-x96S51`oz9L|g*F6Z0yMn-sRn9;84PrcU%
zuvvuYO5uaqAyka%%gdt-opn<cET%7wYdh;e?QB42lgN2QBIO%M{FQ+N#w#yENupN5
zNR07S{_pnR37s-biMbB4a}4#A71{l$v_dn8)b|T1WtLX(&;Y5U2;(bx2X+<n@7<y`
zk*qQs5Ph@^5KCR6UrC^GU}<6E6q3^RDF>zIb?Xt1{ISx}t!~I7tADh<_X1>&6-4%J
zGA#qYHOOUm3`%|rZy{5x#$t>I(2UgEf^N!8erGMmGyp1JwLonCgMIfd+aX}3bw-b{
z!^8?t<Xq6;(L;r)XV<h}Y`6NbZ#oegjqZe$R5g_r(%^<gnu(#{JzDOk*;zJDG`MVO
zz|XV?nhJUXSCbS!J6d>f?0yNQ`jqVLk2u)GPc3(tP6eP3Euyul%@A)8gHFJ4rfWV#
zPE>OGxm-ztX;&eapZj=^RaR<=iJSbQQh&~xV2Q(1*}?4OGjI>8T9Fb%ff)OmEsvjQ
zjs$|(r+vbUiJ0k`SQ3vv$!qnK)3!_pRR9w*E=7CB+9SF4%^|_3h(@Z<JWv4<G<s9Z
znEg+=s}pGk!U0Pdr=v8zfp$e^g=6CF&)W%(k`t-H)R0D0BxLbvq<q)LLKReQSUJ-S
zoZ>Lv6PUEWZL{_J^93V1;!1uB`|HD8!Ctom%R$6tX|L^R`l&T=wcaIB{7=koT0C$K
z{xek&cDb<jeI*478xlpNnhA=X?|>InY<+w3voa;hEit<esQ;uk1yOAWY1|?LTL1uF
z$RXPBTTQ!sTjgTH?1#@rWXS<tP@;)noG##qhEA#)kL_`PSf_melN;RPT{Cyji^G4x
zM!){VOMy5&GbRnAcz{+XC)km9wKq1rKpQ2ZZ!1$@6QK;I$=*Rc#NMF}P->sQX3hCX
z?^eUPI0=!JJDKlA*JF5H?UYv~RlUo+B+9zOl(n~}F-Gr+?pUoJRgL?ddtcNkZXKmw
zHNw?&r1)FqCp%H;buOTcTbMb+XJ_^_q8Tzy<eD||Xg{%tO1O)zx;OmLqI9vZI3ix)
zi5ykCU#_Boh|X)Ad-`DsgAi1GkXjgLZtxze=lL(oJrS*`kp4S=;=yl2_@LNHa9R>4
z*6$TCqiS|L?9$Gt8O2xe_aUI5Xz}Hh9H!U;U2*k}0?RWoKc!w{RnpRHgxqdmn%VB5
zWHVQ$aV&Dcx3iPSf>$a)iGiH!AcU0(m?-9~KTq@&(fj$o1!7K8O?*6QMLX;km4V5j
zB4$K$vr#FV>9YNK?TcHrJG?IfsDs7}b?nEIe}+x1j#}1}az~Yg)9BSC9+@(5o;C?!
zg49;se_?GUQBSo5y&<X9xz1gPUxW8ug+9e2A}+L_uUvvax9&S98muqGP;GmoSasj{
z?S$2QaNYF!vDu1r<<#xG8-`APs|Yp%O17Koexn|jbd9u>^ndU!jan>9RMZYX$YlCH
zVwUo~bYFR!b{5}X?V6~UciS9$)qRCgv6~tZupPviu0#}n<{2C5!_(y&D-B_7XbY^~
zOKvmyZy2*qQmMHGDq_za<Z#6K+vm=m8!3a~<P`i{1dVoao-&H$3t?J8>jqp=l1Cg_
z=NW+jV477Lhv~GgO0*%+CS?hrl>O_M7`jJcH~;l#fvgzPtN;8hOF&YOBGJXazn29q
zyXqy`4}19VCvosC8{<d)J>#n(<jqF(j9#CfD2<-&?2ux!5pL1*x1qzb`fP|>TSf)<
z>Cg+?Xk4bkaZ%&hLz`=-`bXK<c6XD_BD<vFYqRXCB`K7qzpJ*I6iwG3nX$p|W)}@!
zRzky{M6hLs3E!g9yhYC%$|YC*t4T4`Y;)IfVVUVmLZ_{@%}U8kCt`u|a$&@#+r5-T
zGhE}DlI*u%XJ%*hTr4Js$h@0}O6`W;C`2E)%46S3W~!T`YD^o?mcn*(30SBq*CrZ*
z@#A%b){h^r7f^Z$Y-N1>*w!wPW@$jQIq=0pJzZ*R&09{+KRlyKU8=LdX<lnAdiw4Y
z6Hz{#wY9ZZdhGAtkM+wPaBrX7k9r*$gVuhT8yWM%htX`YT#^5J!1XGpoP+G^Rf44Q
z7cx43aG6{;X6}CdX*rC`SM&BzyN$SWLWKl%>Icf6K(YDpXZ7RkA|m%=)scey1c4)8
znN!HPe&-WUAg^lJ(FTxAgzs@T$A9=PeceV~PhxTMP+Y=g<5_C;Mt^6YR53S<e$mCN
zXvEIbB3f@SxA;@ve)VvUN_*=!#g_Mt5f73StGo{@QpFsb8XNfqI?c??-bnut77*q<
zL%{HTbGY2m<@m^p#+=JPDUA7crq8Wz^aWhekI|8G$7Z#3sV5?KqnoIZ7>4v&NlD4k
zV8^KYzu22}N%Nbd7&jbQA7;{MYin0s&>th}<HuyKr#Nc)6q9`kIXz4`tUMeX(U-kF
zC30JxxKQn%i%Ni<^co`_Z9x#8SHf*S=CD1K8}H-e+Uo5yk&s$;*xfni^Vo73RbrR-
zYp^!0;*!To=HW#Blgiz<t<|^4Qqy@!<1yN4;?8YnG_cLr^h^3{y!H>^_RstH3@aWw
zJ3I5Wj|+s6hEj%wlsy>|_5d)r&1`4ty?gg&jM8Si(zgpkB157`6Yl33to4aCy}Dt&
z`1KiKiOZ6n4-*T1u-KY6Ozi@O9wiDlY^6ad!2>zk_yQCesAK~F+%p2~lv{Lz6^lC^
zB>3iYrAt4Tn(<$X8}!6Ia;vk(D)xzUid?@AR_y0|cr*0)x8CByr(%3_c<EAGgZn~Q
zwX`86Q##6RGLr)lU+J<`2;|LujjjG9u!^!!5l)sxjP>by)?FrP??t2JX)&j{=jXEG
z#5{LC2WL;^25Gbw-UvyIlevKNup6x!kncI%-yRyR_An%s|CaONgZ;`KsXol=g=&xO
zymR@Olu0tqfFxQrzq5aC_{rS5aO&3^zuypVZ&qFMuvuY2dGN?ZtaU~PbV%D;l}?E;
z4W^r!lg3z1y`87*?Y-5{_BO>?ERjcdax^8Jh7VsX?1p~IHl11(FbY<5Fiq0KV0F06
z{!RXT>#NtV8M??`EA+G3dty8C@_2n<-(zpc>2^Ig%hfAxlh>(Vl&it4s6{;^^;qb=
zc6{FR7Lg7M_8T+&KO2%AxkQafiR-|xU~(6nOm(r-7G2#^m!#s+MU;>2p?`x*{@Tmb
z>V52E`ZV3y>(w$1w{F=eez6%495a62r)ZLjLCNUyQL|95sMU|brKFE`%(ka_`1o7}
z;<qXJ$UJ6ay7%hqWnmu_(6}4xEG~s`O&yY#3Pd_Y+Sty|HIFD|+}`q7KbtJReYCaE
zw{IUnDddmV__KJjRW}IbSnu`qTEX-ObE`|o<iF3_=}b{8J=a*TsbjHTp0iiis8Yxx
z9F4;`9t9t-GNFhNd>uD+{Do}vB6PwO!^%SW+Z>JLS{^d03rm{jstHRW7gNmw4RNj1
zvcquA0dzv2Z-*hJ7F<S*^YwCyh!N2fTU!@C4cW;7H}9?Azt)K^niBtOgAOUR2w7e4
zM5uQuq*o@+T6^AnAe-i9WhKmyGVNd-=`o_YAV5X%;<GEZ?K8K8DO*}<&Y8A~G}5Sm
zJXT}kHNmeO3OTslMw$oxz$?pK#CG`U1Qo#rm^xYwbrUYUzrR27>AE+kOOyhFin;8C
zG^$xBuK4D{!h+-?f=>f`A9XkIe_kNgH(g%A+i5zo{Sh>09*~r`w7<$UJS419n{%S&
zyN&dvit6ND#5f_*R9B-<?TT{+?<jFePF}@<*DOWmI18_H5+vf&Xj>Iq6~vWce35w$
zp5PK;`L9caYAmUfW(pVEO3f|1Tg$*I!2GZf&nZ!xF6q%LuTlTA)MjAPWRCio>s)WH
z{t{$nwg|$s;#3LOiVz@yGJ1`1m>g)emKjbtFB4%df<bACNz-$|gQQ34?~=wCI5m^~
zdB-PnTNTqS8{Ne2#U($nD)5QDNYZ+Pu02`mo)jOqJtBsMsHUW(S1m=<1cpF8UFiaM
z@t*B+zMeldOjY;_@@*P<Y)dp(8(jv&IhlB~qA0FpI@R8Viv)4}9>QqZN4a@<_^yUs
z4v+IJXw&EaId?-!IV4y|+QQG}@FWa>`Y!jQdDN3!=AgZ!?Q_@*S?*4sf}mL^GNZ(h
zD2%U^`({husgY`r(kEa;=sQPl{j^XFjz|MrUQtDbI82wMhNQB!s0W2eMp?$PqGSlb
zPsfj6q#L)33O&`nZT3WCbhbYwICY}Ic756koQ55*D0Ruovv}x*Vxm9qm#zt4lvSY!
z&G-`p1YoQS1%+(Z{ABfw?wM|EIP!1<*V9Owuz*uxz7=0``olIHnnz!X*AZTbRu1S$
z7O~SaF?k(aY&TTWai(SGhNy(N_~LYHEPfw=((e>$Z<03N0e9^83o%p$FZu?YLKV<=
zjHar-ItFM_^__~o1NfTpwOi~crd2L>?9pUnIBzfVGt&1(bS<=t`UFp%WuXfGubVU^
z4dY<@pd9OxoA)I!&<#wiUbFLcu)f6i#eX4De``_;mX%IH8f0@X;M8qW)Pfz=V2#hQ
z_s|!sSPx&MZ_~LiaWXd8JF%+atIXs9Q)_f?l(K8%=u_8td0pTa@>R7~nc027A$mhZ
z#N+6xT9u@}zP{mb)zb+-FW|TUgwE*$$46twLp6tZ{jqClf+3j-Q49s#_w&m5!&a{b
zDDuqLMhV~z*e+YRKU#fMm2@T(PLa1Q`&Q6|1YOTRS4KvlIdW0cm!ECWC_>{mk8>ie
zCyN#%N<O>v=L6z<=fCKPsDyC&Zd1DuW)hqx*h+NgL0`xiMViK}9-Na+OkxNLqutbh
zNt=-R(w4Bm9&NeeyG3!nX|JSz-^^|Divo8`YvESBIlgY@7k*;qMoFe;4bEJB#Ar$c
z8y=h5x3-8)elhlZOPb8e5+!$h8$qPHr{BguBa?zITU=hoyM%Y$HvF0#>wJ^%L5&SL
zIeE=X_{3&F?;VPX7Y+=`0pwx9v%fywH(Z3a-IboUw%esjYl=9rbj!8L#(SEY6QO&T
zIy=O6zAZ9zdGr_jY`y<%>E?dB>y(%aVRJZ*d!XXoZo1EmW$Bmmt3T9mo(;`VLL;+7
zu2RX!YlEFHGYqW(H)rz;;{uHQ-rEl4qTxlAv2iX=?gkMQ9l~pQVtJsT0Ig;bgp00z
za4*iznkI43gZ_RwX@&iKX1|9n0wI0(auKt3nv~a|yhfD>t;bYB=+_XPP<|yOlPOp2
zjYFO%7nf+#k^-8ip0zrn<luM6G%b0$^438d`&1ydkiq|kq#N4=JnT^$-P?%tRVj99
zdgci@ov{9-GeT|>gUQsgSQ)K$nd@euk#%F#$2<GI|LZw-%7Evcjf88XFNS6yKZ3wo
z6ZG6e1gT`QU@!FD7t<8J8E9+GrKPqjE1Pzm%Mq3Cn`!>>qkl|4;a>l1a_&Wip}xV_
zTiiCz@*i+G74=hR_SEXfyV9jstUV@|LX$!*IZW0{ZAaF(WRytbLlQzFFb^jSbhWjo
zh1pJDam#;DYrDNXvN1ItQ*PEwK_dgdPD_8LF35$V?gC6ZDzd+=j44{P?WyOULz_l^
z|6Om-$s&z}jePymPS+>}k@0>zDj{~T@Iv~?M|z7EVECQV={|Qlb#ZOFZ@BivPjVMy
ze8rVXNh7Hu;U-LU5#e0EEG;T1#HlZ(->S?a&c^)a->D-E+zTfV^*QonA|ivIS^ZLP
zYs~DQ^mT^y73!kqGxzYRvzfG}7om&#J+V<nK)v+Ba#3#{!e^2GuRuSEzNq8ux{Zlu
znn$1odv>ZZ6`Wvwd8Bf=d{;cwf9&%9)5frih*zDLNmyPuur;mr$A&y7eHSrBW|npS
zG%Iro>vIRSfO`QNQiLn9(Fdd%v2bUub`GT!Ay1ArRbnZUCBF2RvzQt!pBrTCnWyv!
zO9%%Er*2#ie)-fWhV&KsfA<Db!(``L<{stPd91aJiE?%-CtJ>M{4k(w+!xYBge%e=
zMD;(zBvw;&DoKa;%Sr?qhxjUrMafKJrl%y=11=o9U6C_t`yRX@vMVOM#l^6%`t&@J
zcBhiqkB9Oq;96<>mO%QkQtqETjf~)#kdWj^A4x}7s{^dtFU6+4z4OcBxAU-6^@lCO
z&9W%AsuVJ8AA5i-5e+th`O<pJv>47!db)x70tmn)k2bV|%ruC<rwF3fT@MPzh)l1T
z@2GeZlO@Tr(Tte<v(;0|hKNP}Yi}iG!KQ#q_D>)Tj!9EB)_<MNvZVjV{1K!8->m)>
zje?&ubNyd|HTeG=>Hc-n{NMceSUjq!f1}j&;X`aiWo2$|ZpR%vyZl;VyHQkodwbFD
z#_Uji;JItk(qlh*daA$8&1G+GZ4DGzcICW#hckbflf%P(^=jddA6neJyd{f^HeIl|
z`T6;R-rjraaRi|9AQ%`J=zV^CK_jTEcX)U)J<@T#+`+-Y$UD!^?{s@h%c@|jAF$aS
zm0Va^DU}buG_a0q2ZLS$6hR_8fxvmsNKGBVk-ume7&fKe7Zvpk4Rw#6i}`H~b9QpV
z95&k48(q0_h1TPK$WP)*`brZFrvJ&_nRA!eMmjGcH9__#R8m^1Z(;Gi@4@c^$`5MY
zlHH7$C@I-P`F%>{FB$8vE;l`ltAb`IovI(cI9j-D*2PIK%PKO|5DZ%axs1!!@4J;=
z`z0Vg=zH)fKYx3B9_`nkl#np<ATKYkv)!QVX<a)rm6Z2EzXT#G{88APtoL|53d9t7
z2IZzSd`4d+T$lSnlCjhp%Q>oP@9#*_cB*;q_wPay7^453a$=Q%s%qqMM>*}~4L4Es
zLU~5){TdoZ?(~?ti1V`WGucJM9&&PW-b{`@AE4a#IeJKfpMU(H<oC>epgTQ>t)rvk
zja@b*A+<(7`T6>nzL`glDS0zbZzb4Y4OIe}FvETHg|99dhm9`d;F#DFd2|IFmygK%
zDIUQh`YxBAA!&QK-rGwiLc?^4PZ@nHKnX>YWUo{1@Gd?+zI>LWo=ese9N1$?zK`5?
z*Uf!W2hPTgnztwHMuug#u2FK+Vo%RbqM+Px%HEOBsH*=eH}}YA*_${_NFD(%bF;YD
z-k0zk213}ku&}eo!&O22VUs-n<g=hivag~<BtPxBxA`GFnyYxBFTZr8(sk(QaR13z
zjn|P#5eyUf0@kI{HXa&!pHmk1@$M&^zC6P;pM~CB-Vo#qq>b?P>t)D@J`O*Q&<*YT
z_lH1UWdjb86005#*KJ!6lnJ7UyV}RMfD48F)GcX}bi0uXXDNw#4^QEB!41xTKbJ}s
zU7|)Xh~*ke2rdA>yhdbL>5?0sfeE0}@)ZT)?`UR7G-+m4soREyevw6F`0MFeF<DMG
zO3Bvd3Z0N_hTsjG{?BzAv6LciXe=8Hw_||8AUly?cnvFlMypCisnZ(C@!%LdOU9l5
zt5B4zVpV{sZ-d_q9a88<XFPO7_NtjIDGO&Z7a-_Qb7is`rIKca#M)4O0aX}2e@hED
zxANe|JD1jkY+C%XILv*Y;y-@8`k+1b%fr?eiS*=wkwku+T%nG!!H$7l@@`b7+(ub(
zv7G<jd>oTYFUjbJ$1*CHJ{huDsXeaoct+)dxFoHheLj2xi9z04H=<M{i&c2e7RM;?
zqPCS9tJ=mI;BcPGC?n1?SgLI;MUeFKK}UIOa7!EOEVI2z+p5&=mEg;C0r#WAXS4rn
z=bI6>yXJ@FLb6dAMNx){aQK;>T8r&?zWNP4{jh-zt$@}DP(UlVB;UfGufMXI(nkP~
z<=<j{e0n8D)W>&fLyywvikx>^Z1i>uH7xx`+=LK(l<&XZx3`k&&(F4)P_pPIZ~Xjl
z$K`O(c^iKKumnk-YLI1s4i5cx%>k${FDgjyeb;HF`J4y&<v(#<9$qV!A2=0i2D;|1
zyeIxM#e^gj5jgKU=+0I5*^A95Xmfv!c{f^S-=LO;nx8!aLOq;yH1Vnm&a2a3r@cuf
z7{=sc_x<5liXEH5fSvY2f`9|oBk0V2Ntuw;($TTidLH#5P|-PQ*tDD)c{aNU&CB*F
zk_H)LET?Jrp;qO5Vq6>s9Zsp~Yo7C&{&TT?LdH>)E1D8{_x0hsCH6wfVA|h&TRI`y
zoQr+3`|Xa4#7@sKmO?CA7o7xu^G2iHigfo>=FG{cW!7r2gqvU&by*s`A>v8N45wQo
zs&?P{t-J;|aS}vwCDV9;!3p5NhfvQ*<XplveFIN<BJ=U&t4OZdOqofAcp0wDQ)xd)
zUz;eM!?HQQN1hF6Ifd02)EED;^XLrtk>3P<KMApTy=;0+3P1Ci<}-bE^H2TXcAlFF
zlpoG-C6puvRDZ(XM~Cj<6DyRdz&}^CMX+-Et=GYkaA{C69G=syb}tTpfqAz64ztqb
zb5zvzxgcRK8Ql#kFM6PA$SCRlbL@CJR(%w>K&JWZ<DbyTl!v5y?^)dr$@muKND-g`
zj`g$MJXjI(;U2ngAzYNdpJ22>+>fkTx(hy;dX7Z;Tmb8Qcu`Z{`93;+r-B~n8!|yw
z1!fQVuzDk<K>cNphYmTPt6yr+{jhW!I1)9>cVoH)tE6i8esjZ;Evv!!1vJ*IIm&G}
zTKcG8zI#V-TAt|1<Hv=n$-?ENVf^lsShDSdT`>1+KAkmAhDbg_o~Uu?G(4_LkQ&J)
zef9Olbww)8Q+DDUe3<DYnX2g8ZE&#7C1zK$Mcn+I$B~Y{paGsNu?K-fvOB}?JmN{@
zRp=Q9R?s&+p=^5c*hl_CNGT(d)o=aTS}z+Bf*?*V_d)GwV@qQ=O|604^UKGTt}7K8
zt>Wq6b4j$+UQ5HY9LWZ#%kCd+bd9ja6-$=djV&(_GDv$5S1XKpZ5z_aT%opoTqOm~
zeA^elO!^ArkMx@)yDjGn#73FtSKYt{K39Jw9ol*kE=;59*VnB5cn!+B<xR@7!nwX;
z3AY)Y6FmZp&Fc?&&{rFXXcn&i;NvF?CaLKK5&5agca@bx3w$N;h3_Vm;07>0HskyJ
z9!W?iFCCgKv#&g)TyGM%mz<t1jnAX}aPA2&Jdcr)kuQ*Kkwh#XBxXxDM;c=UsgRFI
zpD7%mBY;%_5d!oYGv#Yu-uC4S%;Kcyh)I0vidL1y!SAicv8f!k#ISm^P;$br!%bFy
zdmii*cqAhI?rCf9lwTf4))d-PA!oo*a#$9UO`3J>+BKKeaTywefw+)PDB9^vleErK
zPsew)tO?r=CtbJaG$O)zF23ahX$HY>o-Gj`+)aNr=nrTOj_x_>>>`z$R?MuKr(Dhk
zF_)a3gj**I5wL^fmA4t+G;xX`l>8zH^0RMN%VRlN{Iq;NG|RpfR}>4`?yn)u>?iMs
zfkf=W&9BWM?ohOX$1R4u!uSSJg40E|pG>KFD5jHj7<2jNL8NO#JokMAQrr!cBwQS>
zvol$x(1IfDl?$dA`0-Bp&@@HbXOvoGqcc_F5XB&Y$(Y<N2j>C}3N8V+T_vh^+jt)y
z6(_?W|3zM<kvgu>^PcQ-!>lSXb!Z0F3y|-2Iy`6!&uCpwx`d4BmX4&6!Krd)-lUZI
zP%A@ZPRqywUf&T0eq9$|Oc@&F-cED@$)(8#<&Zab-X6-kk|MdbfSxDh^ds^j7QdRU
zk!Pg0IiKyg&HrG35G@pNp8uSRj>PK~o9Afv&AAw#E9c#a6cD?aapujK-Z}KCcjtNZ
zOTNjgMzxM#6u6MMyz|+kO#|^PCLDeA9LDkmbP~?ccXkxW(WXts+?Hw^zxz@5@t>L!
z6!Z$8lmI-n#y{M=j6$;caSFi#m?o=la*3zCTqCn%2vE`<VHTbI;Qu_#H_VkInO#6X
zqL~BnDkSzm)tjoGTlT6DLkc}h>^dg<?*04w;%`&@X`^=px!M_$4?Kkp8ecNj5ym@U
zblDug8xMQzrhMbgw6O-wm5U%AlO7)W7UGx>g=06F<cGqyr9m|I$y^a!0=BP5wR{RJ
zb?zx5c92(%hs;FALj!1}w(h4Xo^&FtUtCtu^0`<IRyyRz-?7+zs6FG`>$~KO|B8vU
z{~ye~byU<{-#03PFi6dSz|b%#q0-U~LntkZpmZxB4MW!eLra&WA_m=!bV(U7NViIZ
zl+?TDy07cLpXXWUUGF;UoORauLl-XBf*<?0XYcR#Q(p!LrwT7S*d0vN_2>Le!p`;G
zR7si7eHCmwg@ho1zUBN}N`~|BB$Z1>wSfMPx*yAsicEL1-%UJX0AR5yObDjY!I~;4
zrQ$|vB}N!PmCyo!c`ZyuQ>yy^uJ`&EF7U}dB3IX#Drv-1=1x}knw%Wl>C0Gj1*x{c
z?oUguVLM-)UFR>m%ZsgITdq0^JxH&QAT;HAe<#D^q<QSps$o(g+g#n!suXE#qQlFl
z8rnD_>M_*=B<v4oG4aiO$UzM5bWpMb4X|Q!mox5*w?Q<ogu#}Qw@Zr=mM?6QY(feB
zSD51AzhH_7NV<R8ed`*e_B7M~qxNdFTlFpytohq~9O(r9g?Nvn&8(^f5-{}9f#DV#
z&$s4YxF7sBuu%I@T~qx;Pw!1KkD;j*$;uOdk2lMf){m9FT~5Pn4;vdAnx<E0xNqIs
ztC3~4-+A+fV_5XgojV7=I?LWsoe>_gh(LQ@Gcw44hUs|LJ}ogbbM25j^M~ndxQ5${
zD};oE#wlh}a_Lo7cVvwL#FyGuvLGyQ@~@|nJZ1d9hHcumJ#8(&0(NBGOf7TT(YCtL
zU1RKb(KNsm9qvx}K52eErI*4})*q?_NIkosdz82Su<5gj%rtp@0c_L;Vj1ZEY<&(l
znFe-(RpOW@L}z_1aKQDXHR4EyG0w*k0G}RnAMt7fpn~G()pp}WpwHRxCLruC002%I
z;EVJhK1`lG!en5zAtmFbCba;WJ93qw4W>W*4agw-lY<Q$86+Ev9+^<V!84UNg5O+7
zbYorPO@G|^{{8zD>ynBUKYNHa?tG1*6dL+;lO6s3@9rjw6aU}M%?;EX8jmwuBEy$b
z2XBX<F-TsFO8ct`S-aBv&yxI>wQCt0nYTPQC}^Kg{*;W{__L_oI6h?-m*YOsk)qkC
zJN}F!bFwJFwsY6DqKH*gl1<?)NtwV&Lw)^3iE;BYPOa=J5?Wr=WL#aH%W{ADOFlV9
z(_m7#mY_UM4z#(<)!W<K@PX2)VU2zIh2Q3sxTMcPQB+=1*Ow&!(8X`#<39dPl6!l5
zXK={VC6y!xTU(TPUfbEH9eM_aGnrNWyy!Hew^~)LUs<yEPIi6n=F;c{*1Qn8vJrHE
zc4mmZ)TfNFrThnx!)^7ykS*L+zdAB>KuY?3qj1n;>AUpFmt{$hmV<ZF&wrNLNc)XV
z%UT-wg&Y($4*&WtQQT5IAaO!yczU~kpd3e9zFfJK<+I!tLc}YwFn#>{r|#kQ>*m=e
zuRSpDZAt|aOHECEKJ)9_ht|~XrOugoaC*MR@lcCBaT`EP$%`rh)rGW}sKfWA<MIud
z+0MsyO@$vzrxYwQRl0?m^_Lnf`zR-??NU<>8r<Hcg1f?Ajjrh~-E8E9(%<j8Ujp{`
z2UNhu>s(4?j+w5SvzX~%dhq06|68NSzFA21;ctW4xff)lF<2Y{-g*CGq|7Huq~;J?
z->9`76(#KPekESm-EH#6kKcL;ie(M1tFb;$5DE*|Lq10Iior0CAqn5UedD$6E_7cT
zs~_jI66ej%hU8)(m=BQLQ$HMZcn551TS<@gG90I{17+QuZQWwdML%G7*?l>T^xs8%
z+U=e7CEu0y+U*P_VMAjO{{d1y87M2O6c-l`hYmZIPS;BA7xQ``VhOtkGLxQe{j61^
z__eWi%hST@agmRy)B5zgAh`(rh_a*eldZ<W8_db2yZ-G|*C^i2&C}*^g>c=ql{DK!
z*(2UAQ1^OgH^n*KRls%*+(dU(KDv^(z0`5*T3b??msEv*Dn+So`^Jt!0jb%6e38}0
zGXyOfiirG=fIz^8j*+gs`+)buJG*ifo(-@Ue&nz2wI;|cdYvmsZERNh#0*QZz1HlZ
zy(aGpfV>)j%g?4k+F*IEH|Gb6zTCx1A~+OowBP7qznxLQy`U?F9%K(hL@tcdr8stc
zxW=c`Bm=v**sLqg;Uz}1@V>n`6siMW?B9SDMtop|A7NW7?KL((zs3HFJ>Uj;EK#|u
zo4WS;PnN1bHb!GJ(@}^cM8`Ebxy}!irGw*Mfi1*iay>qY?ovCKZ>%%LX?1EeAt(5W
zM-!l)9$`H`RHZuo(6G4eh)a<woN>P73hNVj%zO_;loo_Lwc1w<T$5Z3Nu-d~GY!zc
zAm4xQEqMZmpX=o-vCIrR?$VrFeQ6?mgJH_%%Z=VeuROiY=1|%T5fK=%Z-X6B1hJRO
zlrJbmqLS?s(A#~k1fM|te6j~>=0D`W*O9|}cH<=zqpwspKtggkfFBw!MdtRewhLVg
z62eqAg+cmxcV808ClANW-8255TLrp@J;I=)RwoHu3%kyBvNaOJRE)QzC8$^*SEO*}
zb09d33sBAoV}#)XS%dW;{p?qWSyD_IE_q=R1m6+S$`R-FJV_eyBA9p-#L3P48dHFy
z*6pAomJ5su5SY?`0Kc@#xxw_TbTjetPL~u%s*%{E@Vd0D_~+?AD<yr&jdM}f6_q|7
z9yO&`a77R2@oIFl-ouAhxYj`0=itbvN!ft;*ry|mtmfIbn-@Na;7i_sNNz;!<Zsar
zXXK&#*&AGV|L5-~xKc2A8f^Fks<yY%yuJSNb`eQNyv*<b&64JA`k`y`FMYS?f*diO
zfAWj2+)I<V%muZME979#1QC((XN$7uug*>aF1Do4I%H2n{I=VOm}OWz(m1=qSJ+c3
z{JFB<6l!KweE<HKqSQu+)1#8}n*aN}Q-Fg`zTKz%_q*=vSFHui-IIzktf0m4YC(aQ
zHx(qx5)gV-U3*c6kSq->K}5A&j*Q>0g=NtC7U$<L6kMX0dyQN|UhmpV^j*If7ycvo
z3W*^>_p2nPIpQcA9_@bCiVCmKfMDDB{qqZN)xuM6aP3i)4-XGtd~hXZLni%)Q8XV)
zNEaBUNxp{t(=Yg>**1t8@CP@q%lH{*B-1QUHryz8p(ZK5F6%jll#{wMQIN$nKRG$+
z0GbxwXBC?^0OOb<HGT;Ivf7Lc`n8`wcPfI~+T>|P#R)UH0K|6eoxjvJkTjSGnvoMh
z0%nR4AdVTng@)>Y3-sTgJ+<y8+WGa&RGtdYF}~V|Q8fC`Y1Tr{l|~r(9NzTwUt3-7
z?0<l5%+&k-y{ka)g;28-3VBaNgdPsx235(@0(&W&=EBduHI!v{`1;<<Lu?)sEa{N=
zX%ERAe?QG#w&%tzz7s&~?|O1D={Rkd#5A*Wiey!~v<%0*LVsq@zfVq~Euv#CdJJ?y
zPww80aF}lRHdbP^#{8=E#20K->i@b?Sv{<+EwQ_wj}Jyugb#Mcg!P(Vf9hB=X}UmA
zu<go|^7IeZFx^VU4ttbTb};BYR=OMH6Q)T-pw-z8F6N^ZWk`a&zGaW+*zA(Klpu_0
zQGRO453P+>YZ4;1VA)EK<hqBIXy-J}Dq8DfH_pScwd^XfM`0>O8xbG-1bU;ky-tsu
zX%!4ClHBDcq+9ax+C<G3rmlOWu2!<nU1dXlRlRVJE?-5o@lTG8jjae-q2c_Q$|?xt
zd&R7cjSct3uym8?>)(t3$?`gP!-`)%t6<Z-7a-4bI*QE&=V8@TRVO8ws~BM{8ANJb
zMC0sRH#g#&+)Yc*!={(FZ}krxyf|u~vVWrg!sd8Fz(-W5gJx;<`{xStCnnXs2So<Y
zSnQ3he8zbz3kteYm?|TBe#SVG9bV0(XH&+ZOdVrDCNQligz`=(Gs8|{e2C){WiIdR
zd`cH$ri???yRHsxg|-vv-%1k1?l7j7c?>xN3|;GVuzX_4b|8hvpVYX+C#mz8dl+>I
z@7jFgFx`DMwW!%uM#ZzQ>k7@enr4#QFK5}m(gAApJC{AtclvI-=t<kwgqR|#e>yE!
z=FYtKm-ds|2G3=u%ZJWo!)qsJH)bta9F47r#%|XreoH>+tEoPSBdW#9wp4zbnXKwh
zNln(*Bh1^gAWA1RYbVj_pS#_mtc0%KW1h){BtY=h*LNQWu0jmaZQ;TaJtTy_vIO_q
zr(kzI*Rk_pfO8AGF+yT{1_$2)s;#AjYXP4M4k<07=$ClKm^my*C!^~dd4bz)!k^7!
z-qGx-Qo2Du)d+l0A?hgcb-M?8?XL#W%%qH9^9kP<YDQe;XV8Su8B<o(u17|4wm{#J
zhul@T{;rku8L^h-+)dE5O_Da^$rJ4m=%~Q1Ba+Q7NeK!*!&mGWn;v)6C>~xnc#^3i
zm)!Ns@n<mQKwrPxj%x$k_Rh$dYb$H8c_ph%ERjya2}ElEI@tb#i4DA?VXYbV=KjTu
zMHk<GrxqvPKeXrv>nBj2j`H%e@~qxrzf*~_(rxA&OESJP0+QRAgpMZJdyOOK{ZuB%
zApTK>lMWQwpc%*aYyVs37W7ZI648)*XG-e#I7%6UTX~bxT^t#HNu3)yN}Ok+CikM2
zfZ_w0K0zt(?)eYtbtE5BM|-~_LU#P<x5Uivb7a%T&TJAWF#7*fVBp^T|LY9X|6PLX
zb9N8hN5%^DM@0ai`93+xBP8@C<+g!&eSN*f)_G`4iwuAeD~5(N8JL-wefIgZYpSZg
z_74mI0SE1$bsR%=c=$v`C7g4pOJ%6PzXA|VICARr)ECD$oyw)p%otxty^IoMjKi^0
zdiU>}i;IiT-by5W{N%|7P=lQx)ABsF97yG>U0+`>_Y74vdRqyABY&3V<>h0)K52I3
zdm%9~+u>~Ky_%c@kNNh{nMaWLU(wOgby<pM8d_SlfHrGvMj)^K*xdAJ#5-wu`?i<n
zw*Ct@$tM`3d~*!Tz~8{o59X@YALLRh40-2-JW7}cgvEH1cpE`G^cnaWMiuEPkp7sk
z(V3|!`&2%Y8c>*=#2ktS?1}QCV^@EEDfC#!3N(V+m?VGd<<qB6Ew(bE(m`uijmuVF
zPF)TD`t93941;JBV-}TD{QLLs?QvJXc~@jA{uLRj9)Ez81!SKbZk^-?<6X%iV@nWB
zQi5u5{2yEJ5$;*0DNCSc#OXhAT~QXmqMpyTY>3W{jEv})Y;~-WsvjqyB6$H71kmLJ
zK{j=l#js2ozmzDK`SvS~-uv%Ut>gx}@+upE(9WpdrHh`PUP_i9gW6q+2zUb|FMk5-
z^7Gq?s4%aZl<8y&|0`j-DP#ppq9Y>c`ckO%1V^4p(*>)n;5tumVt~#yC^e!rmDlJO
z!WL=(yR)-`lmkDYU}K`Pr&$TKt6SMmQK^VdK_3;Yq{2@?^PLF~Rpa_qk!8J7<BjaX
zri&k=?85_I|AkwCx~6hR<wBCGtE+iM5RVt09>00f0riTb<(i*Im$R#fb<Q$|FV6qG
zs|rVOUB0!3e*V3xPjTWd_(cx3T>yZ02d3DkVk={{{imp?=(ixCt@Z?^IMs88HR;AU
zyIlXH-vfn@XL<$XUXcu95y&#v=h{NfPc?Q<P_C4R!}Ri)tRi8IsEumTEX^3jRl7>o
zXNc^u5ihjf*zR>V=3(ujyE5%7Fd3NUUB>WXr1((CA@}mmoi2tuMJEWY|LE{=2%0X9
z{6r^4dk>)B4P}@Scx&G%l2xpN*%o3T>wjA32i$DgSpEoMB^U=xx-aT&1c-0riqcJl
zIIjYK6eRN{t)q5qAA1h~Id^4p5>ElU_rUno?VfJ-2mh{Fa7T}8CRjR@Lxntyfx0|u
zCX--94&rK~3yewTUtS5>3}GXvIM;!Qw&k4DY-6dzHOXBjXm5U>Zm5G@fwPqAbL@5>
zy8;*SXyW0KJzoE|Xb+pCSj~Ui5C5WVXRi_{H3`BPcHZI;;L?GJ)Mc>sieA6AyuE6|
zYI)Wj!@wK7kk|L;<@!`CSD)SbZY)2Yl#DpDIrnP&+O@bZsBQXY&#mmJH;@V|D=R${
zlP!gtm$}vgD-LWgm1@Z-Vr{w6*dCIC2m(jl_Dz@@@#1R6b~9Sl{cYt}lTE5kil(<6
z_o)G|$`%>ef|gLLlB@WRVxzo^Nns29L2XaRmM^Ej8|fWwB%2NqV7ng*xRLV5`foG<
z<c$xn;Tx^`YLxKZ1a@VFv5W~tTGIi@oTufP%U#cvzT}pdk4lSAa5Lv^<y;)+_<hO0
zR5%yKrk^e4^_i%!XqB%0NBz?Y%bx%ko*^z*lWOc?5A+8dk4)eh$oO7qZ0ZR~I|jq&
z`a-opfi;dhN5u7Ae}ma-$&R_#=7d<={m=Ht{f7Y4IRRwdXF}IQ0Dpfm^fGEKSoK3&
z9p*CbOdyWO`z_>_|J6w1At)NwS$^^Cr?#oVdjSR1&MJk;t!l;`D<JW3SY2)<PMmYY
zn4s$^w5K~AUOSiuiA)Qo<(SFBaH9e%WH>LAFi|vCvuiHsMCO8GMsxTCSheb5`yiF2
zNrzy%#vv0Bv;#`oB3ggI@JC$*g5~h8Jz-QGAQYU2!;qZx34ke+!7)4W+MalmD&UJN
z;DKo0)>6tTebqj>l^+Qf^Wob@w!zmB33gA2D0@4n$+N}on40I)5{HqNDR5X)x=40n
z?|?nww&((zs7syLPG_rf{oKA{a76#5()gS^U(kCuxeL4_ZSoJ3lkN*>bRv(FlJ4__
zHqZ|rN!`2mE@Oc>Aa3~8Zx3Q+<(x?XW$cra8+cyiy6=;9G83IT8nZ+&5Zz$MgVBh^
zl@asAol5J>><~Kzn@v<`i};h$?4teXd&fsd=hY&1FjU0Li#^`g&&db)oD5S`e00-A
z3sCK|i+fz$1?KcxC<kAe8wC$ED%+w_!J33kGCLzp;OGzU-<oZml@#%Ku?0j0z??TM
zEe})g=z9O|T{+HscYc1bF*&_PwWTdXQptwPW15z{^R^EaOzCp5oNCg*)#x}YGYRAz
zYNK~XJHtEp^vgbuvpy7S!d!;Gn0lnH>U1$@P~31U|I&uk+Ji5f_ifTubydqTm$5vg
zbAYxyzweny{SJZ5XN!izNaRTx>Ep<~*frwvu+{57e%1nw`Ul6gim?*?SrIY&iF7)>
z$7wcEtF#7PrwX_FhRZ))SIXG&sc`H70;mmuTFEH0|B3Zunjx@ak@2;)8P0akqvV8l
zd+zuN9&q&7oOTP^4h{~U(|d59h?H?60TP)?ydUAQ3)B_aIjRc3m^vo3!}cEN>G?Z;
zI^sAg$~&`)Rz1lq+R1}Ppc`1en`G-chES!0GVn3Jduoq3E%|WnHTf65AFY*k7-;a!
zw!z5T<ft$=6eN3UgpoGrT72U8HtX`6H?Pb+k3oacbnVU~a7$z%+PQ^-!^xR;_myC*
zbcXzjsw(PrQayR>%3>O08GP~t%RWh|sVBne5iAceKNYgNxRzxspe;`G3dJ$k5qKe-
z-vc;C-&R&#z(YXEO4|MoXcG}n$_n*GRKa<GLQh8x#=(^)KitTDqp5X~`%R?g`!^Lb
zkCUgB02X`Vqt!}AO7=ugxnXd4xOz9?@)M)-ks2No`&+a;WeZK68b<<BmV)sbc8=gu
zw&2WOlnS=E=kC6*KNOM2_eO?))7-ZQz7EEN)V5xnv>pcW@{6!jT)w4w6F_JwUm0dP
z&Mnn@;V%TPDR3QdxBmjgj1j@IP=YK(1|kQk7Hi|@eOa_|uUYBsva4W;r6X~Q02Sif
zys;8;!FWM@KjqIY99{O^=@+lViF#zN?hI3QW;8kZw-j=GSS3J{B^e}?C3s1_SxtVF
zVJ{K$Z+hh%DsZaVbnk!*CtXTHFg6C&e_Y4Um?~9-f3I-r>(`~r?3MoV>y{fj_qR=7
zk;dTX#0xSex|SVN=CB(m)6Yh=kpj(3mg{`_HNyb#Q56oQbKV@D(<p5Ca4i(cA~(s<
zDk*jL8R-T7j~fp>;lSJ#2Fx8DWAKSm!;)ahs@gCQtD3RnRLc*&-P?r<0&KCHqSX_k
z<IgtYXzkXIio<cr`ES+k1{Nx{!d0G=uTnfdymJJ=W|nb@m~79zt1di3i=nJy=XH*W
zM`U!O^ZLOZ#1R7YtpdjBAM}b{Y!V{+69RYFzXU?)yFD=p_U3A5tSlQ$i;4t#`6uvG
zT*c56A3YCW6v*y4zn&3$@WS~UK;PGaU{}VLW9!fUn$gix-I7go@6K^iQo6dd<;%Vu
z)uqDou;F*FllqB+vlXPtH@4g+@op5PtNyUWl1qKwB${NKQ1^-q9Ey$&8;a7TG!p>S
zqJ}<x7BtH=ZE!J)%_AGwT4I@{pW$38vGEhK#ukLoyce1zJCs|MWg${Cxl6A|BMcN3
zR_j%7=@mJRUdw~YC+)hWp&blC*(PZwg3Kp1qhgGsQxdVsh=}?})l-VEHQY#6e6XA>
zdC=7*3A_j)!k!0pw6;J^>w~G_<q8n>{EY>1YX@Bg^e2D%jR2x~u>0josd2Z4y^LCt
z$@XT}ZS~iS6CwJ2@aSDCuL1#O%cfp<+#=OCPr%>ColZ7-md4>l+*_#J?!_hwiuE|D
zHz8$Dw-xym{rmUm7j7=9b-#ir0tg8y1!^6b<jyZC_g_gQX=uMM0;6~5clt^;F)4Yk
z%)6nE8lCsfm3u8$C<KyD-BK_KpDq#3Q*AMgy)@gPYm$OJ{29&NQ@^uorU#Q3LcP}^
z*GVEvLusyhgYx?#+AVD9RfuIV+?r_lChxzB#x$_RRprf-nlt{ECPwvkhYUFeTOXqa
z&GCnOc+AH{SA_pe$DC{m^#92VDR&-oGxJ5N@))O}UoDyjj}KXzYIrf~I6ZH>MMjhy
zky2oGJEE3zQ9(*NVq`IXZ@2d-ww%>V6;mlT#TvHI*^YhTjL;NHns@AcAOmDb7CX^Y
z;Fn%VuehvQsIaQQ8B)l(rZ;L3s+nL=H-a}B|Ct*m(?-wZ_7>%PKUNBbLB#Gd2XkhD
zSmNa$Bl@+jpg5fwy)r&ABCyAtW%nZurVP{lSq6JRTs$Y`ma?=cOt}iN`nxzbuld2k
zSkqIy;c+bAJnj40a(a<qgg7Fm$NPk6<GJ4lqXrY3xQxkSPgC3V!FO$S(E_w5H_OgB
zEqJ<(9O*>`L@ZBxc5RVsJEF1N(F0J$9z|nDTlgzy-|mE$nR|1+3gTm;$lEvmKud?9
zyu|2kS6)UZ{b8zbdghS8fgQLO&X=>J?xmsKsZ7vcrp9X5#A=Sv`)mM^-&Brv3;npz
z38{Y8q+MnN$1&G%25%FyMA69T{aAiak^5Iy6*`yVEw;UzR5+K9?A<axArd%~i0!&r
zqr=>d6Nt0+kA&rq?f-=+Xj)H&gjfG6!CY2v8a)uK%OAS$Nq2`h{ucpW(4Q!c5C@@n
zCsyRmU|oC<{8%FFNo0+;&v%~_qf638d%c^&H50Z)8fnp->|V+i6IWd-9Sok9^B8E`
zD}BLUkC=^Vp<;_CSbz61!5q0Z|FvEJnG`k_5`S9>Ct+gQ11bVOeUTVvgVV4olqcn{
zp%cbbOdG4y=YTIcRAZWTQdSWQi)oO;+dW~IB0X*4j%-a8HC$G7CJEbeS)WjG-VCiA
zT>GNFnr!#!HUqw!Y~p8Ih-wEU7P~^?A?+?1zgf%@qC~_d)F&k5;dvIX72Bg+ee;DB
zb6@Wfo1j(ByGau1ly_gy?j$;Sq1$&9NJRN*37nJ2UkG0BcV&FOyF(i}lo=Z}tkSaW
z1&HE1<bt36-X8DdF6LecyZ_d=nGGC_iV5Ig50tQFD)02pvwwNVNK3n0HM$wj*CqL;
zi5Lw}VN;B@^bKRbYEkw`11WfP!!287h%O*|n+{Hk&=!;<?2be-C!jV!uW*2K+<>%2
zA4%th&O?z_EB@l0Y#?^sd290zT54d4viutQscjzQ;UBcaEGu5PI{}5O#j?Q@w>R>a
ztXLpY0>oOJyX0(?oGFjI_F2srydoeHbkNW~HM$Zq1ZRA}?fPz4^dq$H?i^2~4dcw~
zzmAfBlSJks&huYbX6QXZu<^U3EM81vR#p~sLo6rP%qx2ZH3H_#KR$FC8dE$KWVNjj
z;bQxZUSaoVmhr`&Z*H9ro$s(*jLHCem5EA{Mmp9l)ya!M7i?w(^S|Z00#TL)Hu`ss
zOO~yzzW4rE8pFR*ar`(iy?pn$K84BWC3mo=HBd@T17cTN0VZd_>yXbhzbJqEM$_5D
z`Q1f8&G~5$g*f$R<N%l4?jFp$6_nYyQ`09LCs%!DXD2;n3_3K-e=z1xi24aSJ%QGb
zk$bLAzkKI!94{1)E69vNdHM%xZjeM4vHGlt;FHdz7vVn!{Sn*Gb>bi*4hEKT9ZIvG
z=oYGcfK;6B3Xh$T`bavxM2*s#QR5{pzKGku+ZW!$J%4Z-Q^<XySVlqMg*^<~m-C8o
zfmGVp;XtR(|A0i+@l8eQA0F%558P9;_m?lRvVQ%_DYjWspeYN`4&ccgo)$53%i63U
zwL3LBmBHV4JJmd%*?6Kvco1$(0!c<weV#z2I{P$R3X%#*12s)DD!zi4k^n|MDrD~w
zwM{2-SK+9@7g9?R5%WToG0|e1>wiAN%TMaGa)*!jDyKi(EN10gi(%Vo%F4;I!lCc@
zYbleF6~`&5S>gxJcjc2hErd{h2yc+Q8Bm2A2Jn}I1<>k?m$|pt{QZHa1&7m^lu|h-
zj<_Q=n(hK>zzA|hN(rVhv5{5wXLr|yB8$SD^H^c#Qjfg7l9<vd>CCIYcH2Kr7YQe8
ziJwV7>(uVu^Bud7c@Qo!P$VVvr)Z+HB=dbE06DkwHkbg;FQGP_weD@9@QDM%1O$0}
z+3M&BK@2V8DyyoQC~G9!a~+^ZGT&w0w0cwcL|(II+0sgX_9IX`>_2f_-|0Hx+ueqL
zv~WB8b?;=SrrIOdyPoGJH)>v9g6xRwX%v)oXA?PR$)q{9c>OD5sd}f<?zXDJTf-LL
z`qDw1`Y1rtR>FeC{@$OJ8)_+7;41JV8wN7r`b+mPQav}+vt=^3R2X_W)IbUCF7Nv7
z{lk_S&edh1_#74cuP<_Z?J;gU3rc)yXOra?;Q=Y4#Tr+AZKPE0`m*%v#sTSM8Ix^5
zJq0RuTJCIl2B^35n0h>Nb4;7M>l`>ZO0R=leR_7#ec}jH9BKKN=Oqca4&a|BY>F~3
zT?!Mri<Ez`T2K#poRDf^U}3%@6~Hqfs@^|1RFywve~YL8a9_5T*TkO1!N`JWoUfL8
zd9V4CTNj1BTsI#8mt2Q6nPD_LE3ZwuM387Czh$_Ghh^#e_b$$^XMgtjblu&n<NTmp
z?-+1G3EcRvbBSOMKcNlgo_4AvU#TAaBv+Rfpp>5k-7QJp?^XGtPP?ST<Cl=P_Goku
zw*prYu!__Gx!!WQ#Vi_lUj7!qxRL)i&&yv#JEc4oe!G(QdIOuc-Y33_Wve}hr%%h1
zX=$;+!G^T5k9ad2CG~jA4@HVA;%Z8&r9XVIzk8dnv7J-lD3{4R@fA=$V0RXKir>8H
zzM|r9(LTt)z+m)tsn*&RA);`#)ysE57}+&fA92XpSkL}au(!Z}L^hEYijzdkiBqs`
zqg<a)RCM)dV_%`%X=rJEyqZ)yc(*ZCsNw|2y_G?K=AoW)PFIb8WDPhd`ir^fz#45P
zOqVbuDE1{eHGa?t)cq~us<usfa-mMXfH{Xzh%@X8KLH(NUPr-4?FVzBHn@EuN02LD
zO-SXW3g9Couq)8lWeo6M=t`4m+hvVx&BXXooi#dm#lfbr&)Bq$ksrQ@nAF1jCMtyT
zs;sQ+xTu1Hn8r2Id(I1TrpS-4O=5_gL)&g5Zp?3CF;qcis+k7>?PH;eXQ3EG4Xghm
zap#15SSsY{T)7QoJy7pI&iw>BObpa*NvDjdt=v?9e7g;kl=!oXpu?+kl8a{j^>16h
zS^zj?Q)rWI^GdeRUAePw*OPF%I4pV9V)XbAE$enKrT>l6qtEP)h`J&v6O-TOaL3yi
zsF%X(moHc{>2hWhQZkq>RPk7vll~?Gy?cUyw+z{Zd&F35tI818?l3=N%tSI%u3p)F
zk>k28)}k)P!Odvln_deG63Ye~2XMxJ`HbFflH*QSaLTI+tQ0h3Z%1f*juBX`dfwMu
zcQn?jx^z`u-Eucmo_l_6k*ai0=Gd5MONgxxEfKe4&ao|Ot^o;Is=MAKCN2}ag0zC(
z^vdy_7ur>YJ2j2tcu4`~{r`f*{r}(V@}Elvw<!OgTW$VNKmN6JoWOx^|L=o?$9{gz
z!gucI11=xnvqM)&NP1?@NM5~qm0b(eOF&>$3^caN%F1Aq4gg-k6z~!vz!N5`{P=N3
z>ik)DHuLZ-=$ePwI5}6X(i0LWWeK<4xxRe+wwzN4_{L%Jzui5dB(&aiGA6(tlX{m1
zs9`<+yAS8A5mdO^HXY3!e!7nz|2ljLeK#~WmkYE6pClwDo0|&;`M9{?IPcAs$%0p}
zI@TP$y&EPKv7!pEZz`5cOioCeI!u>4%h_dq6RpOMY?lr4@m3T>^%qx(&EDH}<q^GA
ze;~3L0jw}}(?*7d_CQJYyJv@={O8)*o^vn2_i+N<ljxC0w+$*f`}_M(9lOaso&a{1
zuZPWSy&WEW;tEWW&lkUdH2MJ)NR|L1A4ia2U>rNFucIT7qWDRqqug7~Z?1Nw)+!np
z`R?E($7;4FwihVPNtH7)hcuv9e~h02OloIb8HywO33vaLT0~0++zQ5GxQ>_lsBq#4
zT=at^RN^2)f}AuCr<nuFTq!eBZ5q5}hSWv2Iw!n3XK|t)PL^X}5!_TJx8g^L(Gsfo
z`?qk;t=Mb>M#iYo_qbvj(7K=i{LxLmIxkH~yEF_;0l@sUHDL4FVp-Dcx22_Jbn`9b
zFNLfE9l(dUSiCs74`5BaKd|pJK|y$>m0zdRf2fD1@nWTLce@BwBoYY7Zwf~G$?#j^
zaxMBpyd_Oez7BnpDOW3B6($>A->H0Vx4BszJLAs~F|j<6{xgNHviS>@7}u(Qn{r!8
zX>oBiP<~AUfX9zDG);Vz;PRF<s118=1LUn5(EKOxka0Mg4fWY(j;l?&H|?@y_6|``
z3*c|!_~23e*=B}qXZ_9nb>V_k#Cb1M1%bXi3n$%Y5_am9lcRoz*DBmf1=3Ah-{Blc
zxje5o@_80;4y40}iGm8RDSt28{$)Y>5%M%7aWJVECP5<M932FdW)om*PQ92DZs|mt
zsv0)9_3a6M35&1r-hU=40u}b6U=Ok&N-LK3+<XZ@>p>zS95@l59MyTu6#}E2n}Q1g
zO?M3tTd)w68a0lk06vHepO9jKt2_2zCzkmxDR~_B?2XOA$zJXH#o;%D6UW(?)w5Zx
z4xF1~6F+^(4hHv64lWp*?Jvf2G$$`y*DH*g9^oz+ivlWqD&y!IP6<a}a?Y1x;n)jI
zev>F{$woa0+K0`nsQ8S##LcgdT0}EJC1CBrWr0{PPy(_z@^_o6m%Fd&v2^g4n31C8
zB8d(^bjzYp?5`8VC<0vVafkIM99;49BTOjWt4k<xBJ9y%8r!plH!-3d#O}2X4c|RB
zr=Ef><@Yp@HJ4IGlUYhmoQRMt>}n-FuIvE5DWCRBZALV&LImN|Z+lN;Ufk=wTKO?k
z#9`M}+Zy2Cymbk*c*3Y0T<;8S24gJ%SWly-Yy%n|SeO_VmE+=2bkxFyJ-gQrX8hOI
z8Op{sWqke=eZfU>{D?U@Xao!PX}V`@YE;@I3FLND&E$Y7X2m2)Vj}!Cri_B|hg88F
z76je2*b7v45zrF?1lk6~!^Lg@q?+kA+-L4~8#Y9M*y8H0tFOWgaQ@rkkW{y$ZL`TH
zugW-fcsOgyZvBGUX$aTA2@aTAgc`l#e+Zt}{2<MR?-QwY+I}a$T&M5120J(el7UHw
z!_k4JVsqaj!Z|A6+l>`DFJ1|v)S+I)kEaX?brpCL1!({o7bZ^<-BorFUD&p$B=ksR
zP@4e1RIi%MP+L7e3Kg>$&yx3V>Z)|**tIfm9`?a!>}ijqbeXJ4qmA%B&zR75`>S{|
zLS>5!_ITPbWo3)w#`On->?eEQ9yjh+P5dDYmwM?h!D=p8V#*kN)K?ei-1|X64kl0c
zi2uz;hHIRW4wE&<*bryDSURzL{IO;mS}CIoM<{PWy*b9{hrqF81;E1HJ%Km5q;?T5
z_TAK|7u_B32m#{wr<go3w=%|Yb@U%xyn?y;o7=(Wwx$sNTR7R0$<)b1JHoJ$<1$k|
z1Xug@ZbCvlLV^GbIBSCWSN_aMJOw(^4Av?aW6ISICwZ~1vRu+`qHeN?biLimkK%cI
zjF}_Jz&&>Hk}=Z-u8b=fKsz1b#e5P@`(rwc2A4B~gLgoJNGbK5^708O;?faa;W)p+
zk(3f<S`gWRKrjSAsSm)Ut`mK<bK3{0+jKtsEgyA)3##FsbBt|-h-h{!I$Gcg%+f3|
ztW241TV@Czt)yQXEf{SkM!p3gAL^a?ZC}r!&~UB@U$Y|izxc|(UU)Pt5E5B<L-~@z
zBC@hHkS*}gDvj;Nb*2t>%p0@4tEBj(#GQ9{6Zu*BY4}A{%|@M(n!D_n1d90N^m%P^
z{D?iYf!TeiBFtEvmk_p!L&v~xrw**Fu6AobD~U85^k&DFgSj%9yP{9a8L2{05@J&8
z3#%Ra6NSZDEg`3Fa%kl5PKLxg>C={U>~p3;)5H3K&JYV?(@!78(Bp6$R@(%M`8_~j
zRpaI@F~@YHLU#xj=?S!ARj$zo2+d#)w~@#jN#`I=<D4;`8kQtgO3#OTi-=>K*%10s
zA0o-sLY_k$OUHuAg;OVWZDT|)qnM*%yXJ;3a*$p#B;6{3dmszkfY^ak_A=*juT~^y
zQ}^neRMOw!QtG74Ia$qTyn&RTnT>rCshK|66MR`Id2WYEwy`Mq3FUjdE0*pIStNML
z0yp6&H$8=##IAxNHknO<wcJC@x963vJqZ*b(6K?Gb}@;Bqc)hzk0`XZH-UEh8aIQK
z#*IMSc~PnM_!ECx%~e&R0N{L@eHc%E{ud$H{PL1e?)v*9FvZ|#k3#0?BN&&#64QKg
zdF<!$pS98r;8JN3I|iyAUq2=q=^p)KF!$`2+2Ia<M}O>#BH?~g9y0-;*DXur)M_yM
z4JI-5lcin?qu6p~wstg_3=bU?$W6e@JG~ti%Rg{iSm>i;#yc7WwJ8FxBm0D4i?F@!
zfv3_nQx_6G<|fy-mA~)yJzES_y$-pBtkw>IS+*l9q3=F^+{PHy9ezE(=qdgFPBPAr
zv<u99W<8}|wVLdH?{bovwYFg+`JuRoD`0*4XNU1+6W^WCt(GD9g5&v$qts&_xz>**
zTVT4$mTyy=wd%z-#l_r#MqssM>*=l<u1X+D0V!<}KIyUHCrRa>z&Y%v%OPT6rJ>Wl
zYv4h`c5RXvFG+y(@D4-OZatPULJ&C;pHm*IN(L>7KwcNv1g@GX8sJFoHwzATW*7#9
z(5$EiIdS~e+(dz=m;<@#EOkp2M4ABX+A9Pm4h*)Re~r|RrYD71+Yu{yB|;jk253YB
zq4F??`$`1MhNPBhI<XYVFvLUR?9cGfuM<1WRG=zzd&-spLI$QEHr@jEV36nmW$G|B
z#5}mPH8>ELu^y09TCPUHQ2ega^D_gZP7aBypo9Cd?+NP5)sxYbWWC4Rw0JHJtfqp-
zruYZ~^_M&-(h*M9d;TE*P{vw53CTs4ojC&3DxgT(cP21jIoWF6VknCZ$75*h5h{^~
zSbeMk$qBfm5Wd<jPb)*0^HVQcRpLpT!TU3fAOHkcY4zM&=O12CN4V!^zxa9{`L+$k
zgDNA^@&uNojK4TL&S1o4L42R3URaKq&sOsxun3yxI)T}SG6<mxM@o5MczX{1c)*hO
z{bkO6e@Oz15+qKf@Ro29aTE9j;b;Zq6(H<;*qv$lz7_Eh9!*x^I9@xk`e<u1{e*06
zWJd(?*1*go#X-N~qs;fRPvj*=Ag9HpdnQXJnl_e2cGCiqUsAT=MQVC6J~q}s%b{b9
z0sg?9IHNSuGCE2Nx+Y;_hgEpA%H^rdXeOr-)7x$A7GzCI6XMJ&M1;;rHr6`9&9*N|
z&JZu?IQVF+MwEX(KRcQBQhZGg9Dhmi8@KZBy=`>WB(NI0=DZ8$FYl%6wQEE0+tfIx
zUVnNos2F975eVYsl5DXk(>{L{C*uOUV~p&~WjhVE5X6sy-19<e@18BtwmRXnqsq*1
z4(7q(jy<p^Pi!J6!tY++$<94}iz<Akr-KZdBks4@ev`-v%v@g$zI$4Cy{oLM$5}2e
zxkfjrwTt9$_FF^WwQu*nm43ES<@j8xO4g{+@i6228L+e>-VWdJn(%OPC^&Oldp2qK
z@L8yM1jEuCX&Zd;VRr76E-((b<y|AX+D$6<Xf79O&(JN1N<;)B{DJ%IH>cPgR$swb
zIaQ@gN^{rOfxFQsSt?p%`!cXt?(UE-LT7Al2Xb;Tzec}bSa0)5((P<te+7J{)PPyM
z=1yy?_O_8yLFpRA-BZ&$m~xS245XJ*QK%~{2*@ZDYW4QATpabJ3RQs$0Rz8{ilY<>
zhD4guk<wT{zSs_zts89<NPoI4=1arNK}g<nRj3&w0cSzQ?b274LGpHl+9E7ZGjbn3
zd?>X0>Z*WIdq(b8qfH8&nd$O|l%KLbRZ^REMbjBLXyC!Pg4{z>M<;%aK!Y`{k&nzc
zyR$o-FC0_eW?m5%<v&BLzQHe-Wt%JUO>ieJXgt}A0Vf%#+%!Hp==M8dnKh~at^$kF
zH^(2BOJ#qW%r(7i$glh8U+tK9c6L6fSDoAIQbEbd&!%ZU$*zsKSAo_8&lXJ#XMx5m
z*GZ0A1f0|ctSwkV7_1Us;XugEs<yOXrz6&2!6Aqy4Wt-d$8L^z<B~&IPU?97K$evy
zM101i+&nbSU5qd%ZqL<3Xb2$2Q=O~Yvcf=&`1v5=NV=NiE#_*~aS?I9_kvX6%xB&{
zFC`>i72%z5@p=v}=W|R>(=h3+%F3;llai=l$nB{*_?1r|6KI_&<r8BeikCaCC<Y6=
zLv-ffSm!y5JW<eBf-A9h>Ca0+3Q^%Mj8gX|l41Ce;rn9?;y^^k629o1eIy)2dAAAD
zU^|Lv<>-@01+2Y9ltA2QteIv=Qc>S!HcDbYh=cSQ(FPBLwk^(v8RA4J@YMuK#gS9?
zuJgAk&j4t4>^9eHQb4%jF9xU5?)6XVPxRC53Y3n4_Q3I2+(5)JL%~eyG#mRK;)Y@~
zU-}}$R>taWhM$OHCb&0KxTW;s72^inTH=iFjNgNI?P09r7r2yZlq>T~;%y0nU9&Co
z0CN)dBaMfo%dWpE$R-ClH4YpLOC}yY+Wfwl;?bG{&Y)G=rSCJs4-XEBKx|N^-f^|2
z*4(8mTR}!{h`bGx7NTuod~d1t<C7$Tj(56+r;l-NfLdX@u_NL3F*C;Y?z`AcRDxj1
z++!i-w)hYM()oiVkSAE^*nRJUa*L-|{CwZ!;#;V;l+r+dzp<tzdN!r=5(QiGo=^*$
zfF`t(Xb~y0Iw!L%Gv9(#r9^x)Lqrlk0At(P=X3*b5j1iaYOR<9{zQcB-|a3x*HIaq
z2)+)tgS>EKgg=!E>nc!0@F~-a-c-g{W>Usm-u`vnjUWWig4OU6sVa>WCrgi|_JA`#
zlv~lJt3ei%PD&$2kp=I$Jk;jgs{$bvs`GtrjUm>(yP{64%#{-wo6M3tOzU`?IR3r1
z&C{g1@cJ#N!b5Ot=Q`$Z;36LHd)nj+QJXig3gYbW*c<BG&vu{wDl@(jY~#=&B>Ch7
zt2ym&Z?h#+9(9-l#1Jtag(Y3g)m2XbA-_*XZUDp(Rz2UNqknf|(bxtQwct)H5vvDx
z$X8ah8O!AZHV5jEN9#t2l(Z!k6p2+?)?IAt@4h5?uAY^GAyXCIFef{XS#g;f@tzLu
zKp7^TSQp)DFLB!T8{-jnchI)83?E7bB=OjOD`nGi%e7p&;#K3`WqvX`CR_vVt|t!+
z)1Y`sHO!Ks&X;Ee4Ec~fWonYMnQPZtgvR6Sx-1pT9O^bPsi(@AUhFT-OH4x2s&_>!
zrv-UzC5WS7S#^@8%XDf@=pC?8s9S*r-N9<I{fZe@TM!-!;UDxuEuqX0a+&#MuJ8Da
zp7%S0521bd2u<i+p<jFsHGgSY&JccFi^6{fM{GC}f!+>o1b>9d`6RK2fhlkX$5TgN
zjwWq^in4EAQ6y&%Y8Chl<rI+O=Y<LxScF@Q96OouYx4i#H?EU|ix<^*eBz^xbfWBc
z{^BY#G+@*5@wL!mr3)$Pbz>V3G{09goHX82n9>(Snjs>!ZE)M^Z+JQaAFs;ObY$~u
z?f2AFyLHdi=M&{mD2m>I_ass~nP`)Bz;%GCZRvV0?NIY3_W3}6n$3+X^?EN7nHtZ(
z)RL6P7!>aIC*L~vp$v}kL@dBt1w8o7854J~nBTQ1K}zb-!6^PPhXJJQ0e>mm+-Udw
zjc9E6S>#5H)4kVJBJNgJ1;EVPn9A)Zu89wM_m#6#9t4VfZ%!Ie_&6|J-9D8T6Z>O)
zlEsJ7y-kSSV;U+ZB=nhwQz(VcL%No?v@FNIv__FZ@B8P^lmUD&(^(;MQ*ezDlqYQU
z%|)z*l7;Yp++TKV5nUwjhA#Dp2e;D2@xPH<S`wYhsC+<^e|?;RLc4@-wc`T~Cz(I~
zx=)F85BrWxE6M2dXf>=r_6yH$hU-l-^*zUfzzKyQkJI%x@gGQs|BiU1Xf!8!{VrfF
z6mTB8cq(}T`ZQsyvGi>yQrX%<U`~LPpE2!9p#`V2&~7L{LA{Q^11_p)q38Of1lIcL
z)MQ>Es&@;~<w5h5pgxHzuK)@+(VKl*;-pO3L|__h+v>a{fi`G!ooKbJ@o~!Vf9Gcb
zWP0JpCd==#4ZhN;az4J<>Wl_GyNylvxZ$8YGX~7QvtLNZm=OQcO#X1giFsTI?-Z=T
z=w)Em08D=8vXY@Z)WVCyUJH58{|Q<Mt%0UN3$H8LKeIQqe<HFXFLonI`Zh}_Q#3}|
zi$LQSv544ZE)miA2e_h54R%d3eO#ROfY<8HWDYyj#x-l#)j{X3{B7#TO6%p<JqE8<
zprm{Bl7?%uKHS{c4ok*4NoqVd1fMLd1-9r12*I41_0J<mTuboAW9&Ng2_{+05`7}p
zdfXuvSMBNl5SK_JZdx4|qvCc5TB+Nwd!Z;`*V-}`&1nTZi1VNNnn$<&$pO(;eBGRX
z*kovP5_c}m|85Ji;2`u}*#Cz@!o7^+k8nLWvDq{Osbn-SK_7Ib&7Lz$R+)&9Q1r&J
z;)3)pzoaHgq8ne7Pn4G%!q27?maa!CzH6cFlFi>+Kswx~a_%G}LzP3;@j?)j$0(Wp
z&o}=&lPv?TkDpY1v$bB&&cx(h_4jU@F-ftdaZdz_&a7&3{3=uV@f9z?%8p1g6jmy8
znbT!wora+Z%va5Q5WS<CY#-<ABOEYH>aQt>uD~xKZ!pf-#y0=ge9Co-svToWQ;KzY
zpMVN{ioB3<^Z|A(iKIq^m<WH}EnBNrHh>Ln-C$~{bKf?(psS`%%p*3~{)xe3%SpZ4
zHGXS)*phD#Yf37n`lEKGqh;blTnNvwSS+{UqwXsxGxIf9o;qt^la$tcmP;i(vpKVY
zDkZS&{Ag*YT)l_k_zPLg)o39x-d3vB1?(c#>Z&25E8?iQtoSX3uV((}k)P5t5MBD$
zUc}{dNi4_2FA!z8F)mV25vV^$wm=D`y|{W}?ulpb(Qn14`m#B)Cb0!@;V%&%b4#|6
z(h4Tc@8Qt`#m+N$VBd4bZ6^?Mj60F&>_@pq;qwLgIN|DVJ_rf&*2s(?-obnMchGkP
z2V~jyUrpAj2`D?M1-!~GumnGsHo^}ASeh*X&{NR{c766o%^0F`s!=M(uGZ<szVn?q
z8XO;8|7F_{dC4k~(f^UiLgyZZbHZ?tL+|qyln26tQf4q?AJ_HFo`Mkhh;U0?is0f%
z)q5!EnK;H(ke<0O5-W(db7gG*kiJe~CFa>wrWg==#^5NVMU=v1Dm?<QCcCZmij}nK
zz%!!<4{VpQWD^1m(jG=s^~CoB_vWJ8^p_qw`9Z})ehgIGsDH!;+O^L`wRWT-q7CZw
z<gby0RE~dRnKQ(Eh`!4aytre~;m8)xQ#~iD{9qg`%d?T15t@8p`8f~zllj)e7-Olx
z615^X+l%5^7FE7Vh)?)1&5AjPpJSm*nVDN<RbiE-e;32XAJ0$M@5I}WwrG7vO4xlg
zta5BR70s7O%I~p+GxfXcWrItA{|5z?+gF2Mo9b?F6x3;vG&mt{BT=SOqLw3(AoZG0
z_6NUTnXJ-JpNhU|o4szjvk9OVwWza>Mi^dI(Oc8U2(I~2v_T~!Rr}Q%&a<1P^YzET
zSwP#cLupk(xRbe@`6D3xPERHOOt)2FAoQL%htIb0QoOK0xp0IFed|}<reaN_dsSy^
z2O_Jzs<mFjDh_V8BNd-l4V>j$sA*~R^{kZ@Jg4Hqx4JzQP`(crm=Pv{!vnKsjKL0?
zGR()B<BI;fgeRO!K1TUc<A)EOnvP}_L>@kaAsr*Wf~5nwM-Pz66T1Z8t_9&BQ#3}(
zHmiu}J9cJ+fn5ONn|SF@0u}`u?zWk322hd}eg_{S?#)%IS(Mc-#wmh!bu}kolQ|a^
zfe3~n;^rcH=-8&Nk7XmT-=MJ5NjS=6>L>Q=u?TLvjJTC8i)to<>frKcKtX>ZdXjUo
z7x3LC$8RM&zsRi58Rq>LXzSk#&XB)wEZ4x*mKGZolU6eFxCV%Xmq}Wc{?FC1FS0+_
z_RWq9YOjnYS3qJ>!gET!?9A5a+X&-@<G@t9(nJdRJ4JC?OO)m-Ks&q98A0Xo>nr!{
z5k%)&>HQ|ZOKf)-DJaYp)<8<~PljW(8-TARU6wGiDwu*I^j}Vb(-;2)=^`j(axgE@
zHLcs1VhU<8kk=3^6=jt(echlv#65(*F%SxNOcCo|0MO^z#p&>c{a@h8nkPaP7-YSm
zss@tT?ux{5oTLn+RHdV(t>@pnG}@~sP(+TSP9Q#$*+4CYslQ`ru}A<I!8Iw;ZER2k
z577_M_uD!C%9Qqh`_F2`6EzEQ3U%Dub@9#WA6?}`5B9ZX<XVHGUoRnNjp2sIl_oi2
z1B(3X`OA`9O%{NNuOwp-!M@SRG$b$r-B5T%IMfrVIfge~>zLyQ#!nqYnVh!!gQ-i$
zd$@3&gU#ua=B$Ao<{Qz>TDnat;>?X5K&XrrHAgT>xb+T)>ql~O<#3`B{sPxeq5oN-
z`K>GU6SzZsx3ba@&o6V3?emvagQq$z4|%yN*14Y>hCuq2RY6k3=6r<c3Q&9c)y|at
zkTvJbzY&ngw;tVZO#@tgl{key2ypmu`3IOt%TB99oXjbkorC^d`Hqs|X6T?SOcmhe
z_CUf{39{uL19`Np3W3sXC8w@ipc}_+m?O4-SV$!E(op^s<CDja%?rAYQxcQSK;q&I
z7kbzKPm>!6IX%dlPfelmnRpkd+Lw`o?gVL`n$0LX%7r1X?ITiiu2MFCgm!}dY+~yK
zU9uo1bzLg9_pX76=mK{AQe0gmB+{v0K%6fQ@|rQ#Rhy<fX<T11sRAT|;LulD%fG#%
znBEX={TGQ=xNuo%yZ7dC1t|g?&=&F;zM+8{hg&!8d*0uuLHWmrf6=1uR7I=D0`1J&
zg7vAgnwc*93_A-h=i45K)SVsFw(bkd!E_d7MUNSPn~1ap@GxD2IB`vFq14Q4RBo7{
z)WZJVBgxfQjO|?if`O0n{$(Cf3U1rBx3hCyOr-Op`OpKQ2I7(-L9q@yf;dVkJsh|w
zXPdjBB!_EL=#EI@BXb{-b$_3%$(8{Gftlx8W^ql+H+o5*VjtVDVHR8L*5xu*f0{Z!
zJ^9jNkvRpj19?qa*6kk6)VtcWBbhyJs&sZN@}sUHYNUOSM!F;@y$l>F9hIfXeu^sO
zw5_iCBVRWLT(qpVIxmKO8ykCUWRw-<BX^GcEd&Wp|M{~;<gJg^0fqk7N9}cqRw$bV
zpVw6<qk+g>2b^#RHJO23ynpM~P|oiD%E@O>O7Ee=#m2pK+sy6VrG3C09e1nPKStV0
znXO|I4RIm=ji|eg)Fr~>DP&1?)~J9AFG{n$Q;t&+dsJujDx;y|P{=q}E0bL;qKq<j
zxZ(A>LY(Kb{COb@(d`pmt>EzYz=`2K;727@uIR&wd`~4i-{tz<^B3P)4romdC~NQ!
zG%*S35q*#UF*JTP_(!D^#MBG>!CdQs{id?-F>joe^}+91DvR)A2i@>*Kp$M0ErZi`
zcs<>_Eh#iDXdX82MkF%DfVyT32tmy(2`gb?Vcz~2Es)Ru$DMZO2NiS}(m^g6pM%Rr
z%IENcPC*xY)s93wC~}fpy_jHIAjy)q3Gtf^gV;3Vbe?r7P2dvB+5&+vBMYaDw}u9G
zEXc^oTH%b=)}195VX_t0Tr)Y;H$rRtaJ}BY1X1A~TJE~n=X|TAeYd*1m1!k_-RcG`
zxL(sx7Jaw<8|==94<F9HBWw{bU1DIuFuA@s+i2Foa+vyrQue7?F4SLlNX4LpyJucv
z?+HS=f}5(bTbGyato~mSxBm_SafO040<H2t?Nw;@xqRdK>(O)6e9hU)Co;*0d)VjS
zs<E$9G<`mgSsi9hT5O3<Wc|@wiy?E0ballk=SY0N1j$n#633AyLEc&W_3aQhUXdr$
z8Uzxg*)$?&P~PCqTBZA&JR(Q(tOc_lC9ko?o{&?hAo<bKP0VqO;tQFVVBO8Y9d~H^
zI!>Mx$+cAJ#2hPbv}12<zpWzXbJsr>$XG2OlBk)-h1Qhu4}1Fe?71%IU#YQsnA}|J
zmF!nVO)+xw2}tEd{B#4MQM2J~T;myb!Em}x69S!<TaERJkf*B|@mwrY?yxj3B<XSk
zoBzUKU-dGhe9?UE!r^N}F$&@xy1Zl;5TyBgHhC>Gx~=@buGN2(MfWIm-&9wBi_wwz
z3UWk`hq8WY<A8;Qa)nn&i%wG>x_MWH>-41ZO(om;`OQkU@~luq@{je6S{&oK_Q7+b
zx7z0*_ub$h^DuxJL_M)7m%Un1X}^(h-^X#0r2M(jx|hz(*U3eGtjZKJcWDd7PeP~8
z=CQ!tk8TI9DZd_<=&yiHZ=~fuYma4;^yPFR5J#&U_iV~9Sw3>@)FF@}7y+7wMy5_@
z)Kb&bhom+kO0BNL*??KS;>XGSPa^E#DGathf0VD3;GnCr`FNe#BU^d-l4Kg{I<Ef5
z`4Nv@F4;x_vm^K}M%pXFX!-BgJ6C}ip@Bc5jip4AISbMO>HghDY6;6JXRjYiigNs-
zAildO968s&6?N1vCx*>ArTM+uN6GO{3rqGDCgO%THBKzVWmDCgJS|N-Ej+)E6WaYy
zdm1nbVp3~m_a7x{Bd(gRno9mJvd%gzs<jRGA_#*LGl0@Hq(~z@^e8AGU=UJLf>Htk
zgER~<v`9-zC@PZD9g<SQrlpaNp_}uLy1(x`=lr!Vw$~-jnzh#Z#C`wnCuL8@(CwS%
zY_OZb|J%Uu__vEn2%EZs<?wlO&9g~^FD25ud}4w(Z8Gx(TQ{G7?`bb838fy5EPXVl
zQJP<}%wf?#S{~$kUuAY$6kQHBKBIue@e?15tDDD!FtL!g<Ca8Z3<WqG-z6zR{T*t^
z2Xq@-Ex6K!818-CiphPBtM3ELdfyOB1TG?Rf7bh8I%vH_r7oJg{1+hT+e+-!+`esk
zeNQ~!!4M`x>}k23<+e?+B09SeemaPfeC@Z-$MOmeHC<#YCn(>;2)$D~@GP&LW7D!b
znG%bcb)6Q9nWcLWxUKe(KMMC@ahPeIulqO`Tp}Orp=%JX3~VOxfeualv>##;AMH!p
z|5A!gv>z6EAFd8*q@_O|&$BMH&z~%&*La&-oPQphv1I<_1r@R!8$3j-8IaP3)i$Sy
z--q55(uUps{OOa&8@z@$cZB0kOlmrePnUJ}fV5yITbU~@4b4LB9U&;M#6V!A<e~Bs
z!rZE~xZ^Ajm`2><pgZIXllf&~S(%V*g-3N#UATc}2}@`cHG6W`^mPd8`+j(kb>r@|
zOrB+1kCV}CO)z2UI|h8rfn##?kkO-ixU+aBD3BKp{;P<g^$}cciQxZiYgO`o&u)dD
z#)Flj<*erN1g)p+^P0kH(lx`qf{$H~rHIYaDj`QaP<FliH>9Budu2_f3%7if29y{`
zVHanWKD*tMZLA`$x$U8J!CbozoWnKDddgH9T7#d%yd!i&a_;7cH@^E_NkiM^|F}h)
zNd#{SKL?o$+C!s<jrceBM$HjYO4kO<Y{Fu_2mjfBJpJ+$^j_EK8z2|hd|}h9+I$E7
zI4(h+WnPjg423RAUBLNB7yB{=hAvyksG{%7ACLh5Cf7vv99LE<u6(67aEm03+cZiJ
zzJ&c<Ha3$37~bz9TBN>hF{fmAN8ef}QqDVx#b&S|OASQWfOoy3wC4?|_z3q$KoE6x
zYmJW9L=7&R*C1TzES;F26tfQ!odVF}T5*qTvG-xTH+BaE`3-g_OU@?YvE~E{q8c0Y
zn5lfD)%`#PU<0@=Zr4kTq@~`Cp?u#M<F39U#{6%7hbf^vmARoNLUHnOQ`l9^)!icW
zKTf$wKda#0U`WW2<#2HFF1w`8@Jk}r7)=)!I8pVjp<KOLqvRONOg0LpRFh&tv0I9|
ze7wA00Yq~YKsB9?K?wPD5<8YINd?FMQreXKedD74_o@rnrp{Yh>M7Wjy1UepuI_gc
z%Et6`-8&&vgv6nq&U|Zsl?LEAyWx>OoBIWl<O>#GG^&yNK+yq!r=>W*q3ppF{%`hw
za8WmxDUUQCU~9gZu?W$<dsd)&@=htp28d49NkcSW;bL<1|D?dE5dIa?KlKUQxe4a;
zmZ5hqd$ewqZU6EBIGZKy$1#>58i6m0ZjsyM$kui}YGi0=;oVAZ-yIt10tc8Mw>KLa
z?z28#(GN6hwm9VrK&Iey!+_e{DWfQP%{p{O>+XTHwJ$su?gw~Zqn94Js91i{L_ZMI
zXf&hRxKE&kyZjK=H@<%`wHVq9uwt-tiR0rf1SD`(;2Wr695lW(5x?%67;AzvEk4sa
z(|3PUdzk=m??)E7F|n9r^v59ekb^KFH>G1{0xoZV9qwRJBuw0m)FHkrYIAumYL+eQ
zJ|`DfYh<U88>vaG)6De{R0rAa?;B5%9skcQh~mQ`CweC=hM0f9Ox8kGMS*dOPWshs
zt)ctClU>`CL2+*KTJb1fE=vbrTZ78BQXo{exOK^U<mgm1t4Fo?R-xmV4u2ttCMZp3
z5jTNzTfcyN0hqy1jlABK7)?}j2Wt2$B#N>yTF|v|q18V<1Vf8_z;p+b0apHJC{)V!
zMH~e0_hYGY?O)}2*sE$fnkNM0#DuWlPMB&}nq*<_Z>l)MEgk{Qs1c$E_@51{2b%|;
z;HU0Q$l~4}P9++vt3Roi8c(IFu;qQv>Gju^itpj{xtW<gKdXdp;0x3>yItRMpUG}}
z9E;YbeA-2bFf=qglQjy+w@z#Gk7u>G!yteXA%rd+9Ups|F>Mum)Bpr(obqsmqR{|6
z#bgXb_Px1k2;MaPp89m>ogJM2HasHuos|z(M?l0y1}5*Cm-iJ&xmRYE8XKEYep@v#
z5xwGVV3Ix?{neEu<FL`}ZM@ERbSaNGiH-Hm%*Y+Nh7u*3TPhYZnqH9rEH(nXcc-p1
zWwTPIL=NLD9uyaQI8|e&k0`VwFN6l=V6HW+#2<%Gl5b2cO$aD{ry9^76{4_(?dq((
z27{;BuXQ^g5g+W|pFLO9LgK9|*bsdsT6MHRj!ykPM0;s)6W_m(KT4vOmpYR^hVDRz
zCE-2h<-Yem<&zsg+&x}3ajKFD$p&Dr)nCz^-d~Dc>w$sDMnPdBRCD7Q-BfbgBn*$#
zcOAU?a5e*J@3U&0(D2sXt=_soQg=?j2E3$g53r!$1Hs3Y+&bFYPq!C(_V*<TcLTSR
zfq^`NLDIA}lDQ0|6CKedV$cekN$cqe8W>b!eC#uhON9z9xe;)5>F&-L#4peN-g`Hu
zLGmulX%}5a)zHI{9+xj={wvz$EFG^hf_+Mvd=G&?haWj}sS;#I8Gm+yv&FcFOrlA{
zHkodbWarC&&pH9hJ_D8a9=gm_sB<7ex1)5BsZc&V1?0K)I<Adk!s`opIXRnN%E&9A
z1jfNhv@X&AK7hemjRkiE|AFT9Huv?7*;8)}xh)>2<^L=`R<|7-?6*AiR~nSN$>7eg
z2R+C;Q&<U`w_RD^t{bVTmJn#&GL7u_our8TPRVs|7!*zh)yQ&lCB`s!v<Lyh%hJfK
zUdrgDiM-FyA|TuwE;e;Wi*k>o6R}d|+*l3qI@)=L({Yk>LN}l?&xl~ho9y$VM_b6R
z0hkni^sj}-dJev^)C;Zm0Uqk9o|krD?A2c*%eSuBE-dp%S36{TmRppp89aJ6&v9v;
z_AU2;5y~EJw3XZyF$k~}Lrw?-r#zv3$~yJUO(gd(8c3Ic>LBH<3*R!MSFSdrQIQ5N
z5g$kzxm!((**n0}-Z>t1O$Mj)1`Yk6A*u>NivRzVW*^9J?*P%uPw+qr2>dMEu$w3Y
zS|_HW(@+20P&T!AMMU_ec;nWh1pWS01BPq%A&J8%OXKH!qnhPZKN3yYKhPzpAmE_P
zxk0B<S>6`=TPSkH<%=2g(}MtLp0lL~>EnDTjY-G2q29t{?g#$t9@%BEzfIVH;)Q29
zMzEkcFAg4p>aVnDxa5-oXc7v)<_L<%71?%W9G03rie=QTYXk{1Mb4@nxnBh*hp5!&
z3Ics8C(}w(SNym*BPoB17)Y;QC)72-r#STM>nKt9K3?hk`CJ<yTql%=kseR`+0TBt
z&WL*9Z?(iH63^<e1M}3M8|JS?;?M2!uKVuq2W@eOme$j*r_z1RW9`B5Yye%=gURPg
zxq6a&XP=*_t14DuDZX%DFZdMjZJksqc30t$6x-db1cOnSYyk|%`gM#-8Gs*(@93qy
znPk9*4b6C46JCO(p-UzKsUWhwKE1m7BpcG<<7>v^13LXyP}N%iy2`$&P<xPOP#I|X
zJkZaHurDhG<r~;;at*e03$*Q8YilK#shSeQUiNQsXd7OYHrA?jAA0|xM>Xfd<#9Gl
zG|E2Ex^v&-pQ+%XB}#53BZ4I`%da(1j#&ZYon1#uxATZX<z@opa>3FM+#Ks5i)3NG
zEHmsVaS^wc`@hhsdR(|u2SuF3TJmyd_!KM}#z$;x{FV4$m>SfPZIO_W>8n~s=F^T2
z_pPlM+`B~bs<9U)FZ2t>1zUzrLOF?tZqu9|_ITUjlun`cvqZTh5U3akc*$&ae_1Wq
z<U=i>fxc&?9Hum9K0~1uAq?1NMr4Iwi;JCd&1dSthZbqN>{M(>V)#1E?21|BwpnC#
z%h9LcQLE<&nztT0wIUfC2P-15>K>NR;GbaHEnDG|ot=Gl97ZkDgeIuNR}=<x`PU&5
zQd+cs4sVLT=^E7p(?um!L(8L;2wtYf?s<FmE*pSJeJfpQNRsji(h(+TG73qAm_zJl
zW$2+Bn>^;Eo!F!$!O!1cMPt$-Dd=a>mesv2Brhe%H&{Igg{~GxE<jw9LrFa)^8OU6
z@VDU60++8@Rd=$u%DZ>e{m7u%BMI~A>u_D)LgwH<a#PZsk@xOPGm^8egJ3B%90|W4
zsE&e)jXYccXu(=LZj&(VJCT#{Ax>bQaiierIZB0moSjBsvXY7VC|Pe#P`+8f<fCMZ
znlB#yu1wI-K5#tXe*IbF_5#0GL>Tm<+}^0Ze4<H5>-|-TS$-`e%FpHH%koW3eemXC
z(&S0Y*^2jR{$AQyzjx930rPq0Y2-OpZuI~=J2)_KQ+qk!!d+U$CD>g(V~jHwSm3oW
zkf){tZ3ueCg6=*7@~*n2aPlj=&)<*i{AYfNiDx;wI%lru;8mKhb0}8HtOp2udM5W=
zR%u4gJa`C%qPs2;TLXKOvo!Ydw@gZfV$g=;bWPHAKp*wWs~PmGX_eI6MoBi~#XzQJ
ztopLLoH#j181Yww{7dXlHHf7^_uy9RVBlwA0eyOQ^2LRtCoJ~*4W3QIu8~s@9+u{r
zZ^18u%iG^3+jupTVVONPeQQ8eE&mJ9004x9;f{IDy~bG__8{ZBwHuh<ByFJ0!nu+y
zIbAZS{*+wb`Tzd1Hz4?u-<1CbUtvad^6=zRoEWhXWfTnWs=yB%J=f^|6WRw<3VKPD
zks2<%#()AFnucQ<uKlbU!U=V8J*wyN-r@QjDpVFgM4LvjbTwdLmrvYbHxEf?jgdhY
zN29u5rDM`;s;TOVPYz!1Yd(d;9{n!+Auz^2BhVQRls!LpTM4OZr|rV0yZVT_&3tr2
zusmDyv6D{j_xIVM;O*FRtM8*v5GU&f>-0Z*z=>h9h6(_#tK&&XHsmeJ7w%W&Vg)QC
z@eNJU&bOEms44h$qYQLg@ic|ouvPE^<Q*oD!P3JDF`1;ExOw{)cR01`pE?}c2R|JW
zYDwiv0iF@}gUxqAAF-I$+a9KD_=L<PFK8~{c|VBQ$FWoWX}5}IZ{*|q$3r_>d;)|?
zY-px51Dokga=BWb1e!%o;)&6|;72^9d5BG+$#pOSo~UKsjEhNu=1-qKHZn?G&QsC5
zZMNnGP_HG<`+7f%$*P8WO>u<YFZR8JxBG3-Pu7RMG49*%xjWR6!wt!@#2ZaOz`rp$
zLwO2KT7SVDPTi|`e!H%PQ+9`zavs0?)}nZ;*=gqi;}1Q0N(X_Rfc~|yO;j^IsCAiw
z00^)+U(O{w$j02J%9Wr8)vB_^pJ;65{yBcMRexwS&LY%MCFRpTdj>AgGc?xV1CT%O
zU!Fe>FoQpLBrPBUb`y{7OdQ*I7jIN3qG5Mll&Pr%?m(7zwYBMXz57?t*2<{B^dtp_
zPCzy3RkPHDqf4c6&_1c#O!P~kaLER~%NH6Z;x8t91GT)PZ7r`vap#on{KI}kyPVHi
z^M56>==9xyUrrYVW0vCCYM<X}G{Tv1Zv3laf)XG+g`9_mMOctHxPeie4?k>kzTzG2
z6XAjipqPwj6Tp{4Alj6<>F0ft<S<_V`l<+&a<39u(85nhGwu};QU6;rU6g+9x=T6?
zG)h%_<$e4UpNpOLQ%7R~gdx48|83QAZn1rDpy{D7SWLJ)J&GCn$UweNBpr>51*`=k
zSK}n`N*BGZlwpjIpk2cE@w%)PT^mxn3iOG^pq@IgzH*Isvl`nRnr%aX%<!%Pp{ZB$
zll{*EWy8;)Y^Ie@5SErtb?<oPfqt{%U_@GQZi5vC=$P8M>TO*eJDaZ6w*U<6x!i;i
zqkHa_=E0J|LN6h!Dd48!b?j#LEJ(3l+p4K(F-QWdjY?Geu6bj$wVJnvo);TMPkZM^
z3s!Ov%y&;;Js_m_T?rZvfpk1cAl6+a?=At>RO9%V{b;;2z5m)mc``1Sf^bUMBQGeN
z+vpmJE|d`D0DIQ^?GyUTQuO$j5d~6;3*$qP55#1q;L}UK>23193Z&u){7gY^R&u`u
zoNy~DwC(Man-CH&ZY8CL`XDDxFjDxFO!hAG!zBN>_wzbw*UtU`cV}%-Y(%Cyk49S>
zYL5*(6;Hi6dfFnkPs7*VQ+#+Rz}fZw%096z;*^fken$zwjce~&sO&iHSR+0#Um(<1
zo9WAEl=B(~$b2QY!8-Qu(o9DbN5!w>w{`*;l8)aLK~N+UC>hKbfY8uGc7xPD?TYla
zYXaO~zi|YJ)v@$^>kAp)#?9eUfhz3`MAtRM{R76@6u|Ua_Wc>1;LgvTF%Q>fW8__=
z`Y1a>zG;p-L*(KTxU)So*_(So%`pmQnz+r<nN;vf2<q;6&xQkVY#1BI3>_M79yCwX
z-)q=oVXZd;TGJX2TD5{}*Xj8JCWZQ5-23;E{4U~$$#A#hp#mnI#DIp|(SJs(5QOxA
zB;agv6&3V5R`k~VQW(u9)DIWEbe`1E3MgI(0dWccYlLrJ-Unu0{gdzou3!W@Pb(0Q
z+y9m<{yfwKdp2dt^*IUvTXFezfYFIJI{J?_^@bd%d?#^ZxU#n&_y$LRTK_!7AC#u~
z+;plm*eCs@ixEt<WN^z;>xV@n8g|-&_6H)2<r<4li$jY#r63ar)Nf_rT&YvTI6|UP
z*Bk|e>P<cGA|zp0=>8(t;%r;h^M5|oQ78a;F@r@p*Z%5=8E#H38$~SB0iWu#6kckz
z4&EfZM6GC+R+Ysv#1nGmi=#>57w#lnu-;K+8W36(GJcL5j<dL0`pLsLG!xycL4D*@
ziysN`&)sil=bXl3GBAxN>SIV|<|DT}#u>pqq(@^@v{g)j2}K>{9-gO2-IX=^Qev@-
zOqrODlRT7<Y1lMgrRTBBIbVAk$-f-}?eZ6Jv;Q`RZu>(OXgvg9;5BPhi*0%+Ft7`$
zkQgUlY4o|IK}*O+VoLMmejpwbKD87Nq1Q`;nECgX^6gKe6utv)cn!(I%mJv+U>jtF
zfcX`pIv5c5_o{bH^^J%Knu{}u4==<&Uv2)PyqA;xlQ2m6L?-S_n9RX!XkFUag_Xy9
zd#xR$k6m!>(Sp9$y|}KVz^JxU1!I5RKvcj2t0{x0s>;`6$cu4hnk*!H0Tsz9)hUe9
zd)yx5f#SHw?)SrWCJKWJ@!4-6VlL-aHg3Z%H9-lM&bd|VI*`U0Q%<uZdat9`s#*m2
z_(~sLy<*!n6M#!;1e=PviFcBEiTrOlLs$s54E<<2cCQ4=MyH`XfAok*hG3gZ61cJY
ztS=nDz(V}_8XVlxWZ!PTFtE9TwA}&|t>Lf>9UD#diMhT3e>V{S?Y&M|keY<gP*%b;
z=9<1kgyPp{E(Z%RkZ=-S#H68V9ZhI9rzJal(D#CUA=IRiP$^zQk?Np$h;<fhuf->3
z`?a{iLqbgwU0+}Lb-kGg+-oVP7(6JSglViNSH`5)lC1Yv_o?l`xtSLa$P!bG=w1-t
zP0v8*fQ=}y9jr;_*?Eww&_2hsPjpJ9NZX@y`Sh{i?WR|8JH+U2kvl<ccP4jvG&IrL
zNB58qGPofdw=c{c>;%zNt3kH(P$_qmj+&8;3zp#;gBR!D!Z<ZLf@Ky&Cse;SbYZQ$
z7sn3o85kp<0`b%OJ?6;Q-<?rE4<*S?Wm=~yox&@;pxsUOjc%LJ`Qc=IU$6`A8K!8b
za{kYq_wVx$(i?hdOYZskMv4=zpgaL`4u?u`n~_XH2-0M+U3Oav6U^A`g-o!F4;Zi$
z@)~g4c~o1KV`5l^%1x%4kLNxxgkNl(ib6m#l2rlM=OqmdLusXDic1|fQPg$kpnG$N
zHh-alG%`$A**;Ci&epX@3!(G@|CH#I%oCdGZxe>}-kUr>_TJyYmCD^|avNHtFW2QK
z1GwN!VPJ;1*<LSBY^atFyLdj!d_D{&i))_xKC?g0`_ckxg)r_;c(s&6M12&#wB+~D
z8gLnCILz~~+x}t{rAdNte&sI`6zrj1i;vR_ptMajRaNQzL&ArTelNe(<3qZkg0S2i
zZtjm3lye#4V2vPDi1y`xbNlNP3T7SXkv=pflp{gHL?y*6RPyOa97&AG>DfWCh`9z*
zlLvbTZTWx&q-MDJm$Oki^xOq-8~!PC*0Y@4LH5dn_f?(lLbPo7uJlOTB5n-Mb-<$5
z%Q_(xB~IJ$<^<)<B=MrYh6Xlw@p~!-3~<I9ncz&(o_|VW2y1OmSU{7A3P%=sA7v!%
z;7l`{KU9S9CFpLqSzifHDNUIxT_gmn8I;9qIVQwiJ^gTKLjf6`1r2nj4_0k+5tEIY
zW2uK1&4l|~5;s~2W5|QJ!Xh>TxQK@o?_7_i+$Y+NuMAb8Y#in6_?&gK|KpdcA%2;h
zvKxg1KmIXOVvF8hw7eWHdSkn&*u!%BpU`WgGjIHedxR33=jczbokCAxOiAd^{W^WE
z0qW74Kg>Nfk8<P_dS_(@i7d*T21~LK7oVND`N&4O3!yQaQqgX_BETAa>zo9VjtX;R
zh+y_6$sW)+No8e-o+Q^wx-%BM1;~Dn8w)Q~;pr#6qObnK1^!(z&#!}_-3sfg*VyUb
ztS~+;`)A&Gr0l%&g1=c|cd>6lqGjcj8oKGjw|i&rX&RqV*0Rj7O{5|$e~J)3$J}@+
zMKXnJeZ;YOhE?NZ*|^{_kYl@n#)(KOsMQb*8@N0%_j(pokv0vJIDlR6;4S2P?cUf=
zC4S*RYeVs>=i4s?7_OhC>ZrpNcIMq3_z9b_)~|`9Ita`Jfj!J$q>qJIh<16bYArI}
zdQC%TW90O`*a|%z*6Wu#0qkov_as@~mm=f!Po*h0c#NtgFKsrOXGC}f9S8J2L#jz(
zHz#lbvkd8V@5q}}MQWL!WUP{VANcQwuO6IrOK+fZ_LJnTp0Kea?nx4z@~r4m@G}dP
zhAhf6zR!C?#Oen`jQk*b)XLwA=ws{f+JG@1D=RK*uv8qA4(uxrAyQYI;}>vwU<o$V
z@Yp{Yi!6#{H8z~16SP+qh>4o51Myh=Zz9;-x{qxZeI$a1NRlW7pEQ5o#(7<m4x%+=
z-7+B^goI$S<mU*J`(xsvC2@+Fng+;bd2_sVIp-{RtkwE*iM_pCVZJ@0=xw0bN@U4e
z^iq4yw}JUvx7=J%hS{gUxHML7zDy74%kS&%FB7uLn0*THMl4SYD<;7l5+UUqEO#V$
zRD00PNd6<g_Nb#q5+($;8w-ldD^t~w^c`1RI-L7eyg29El3ttj6n*&a+zA%X-EYO<
zg1eNRb&*$rmB_;Hm*4_YbLuAMHOT}GFWd_b2~65)j~@2fTgms7Z++r$es(HbfN)t&
z=&|uQ!M4i!grz-dc|ZmG3ecXe5tgmGNDzP*ao%Zs_M=9rNm4?NNt#5$-?AI*Po5$P
z=9M`f-Up@Y#hbUkx4oEjJlC|J?*k0c(o`?$^Yn(=^1b@*iO*%V4JMPPo?tP^SCnR}
zZ@clu`1uXh&{iW)cg4Dvnu{p80I}5Ch&ys(w@tUgvvtUr5R0ny&7m<UKd;02c>Xxy
zDCNo42yDLJ?2iQpHKgXc2AnGag1j&<vS<{}Iq;!(Z>wrc8rXrfSSRct&f>$9U?(s;
zW;{{@zl*NCGNAt(U^-(PBzfJhmKgEW57Kb~+7RumIHK#<irW<t1(7f{sdrs?EvPt1
zz!3%&ps7~UNS2tUrrOXVj5RSNA^eOQ6Pz+n$Fz2V)kX9#QA(tyfyYQ}T!X*@5apP>
z@i~qlhO9GJCs<}EP?!qufSi;smDi3ZE$v3_dw!PMk>kA*!nj=(s3bkx9~ToaG(Vpt
zcO8}fr!zN~J58K|Y#$W+r?6A9i4<U=dl&kS7oRdc?Lh8);T$D<#C9Zw?9yovTB05a
zPOW_{j6@W)h@OUqO)UETY@{p9P3pc7m`$mx0St3NrPTV=Rlcv(=#@`nF#;JE{a>_1
zvok!)scUz90|S{_XB5+Sf5-GSvj-pGdOJ;5TJbJP7!@|EiykUPo6bu3X2#?Jr1=g~
zjnD^@YCu&j<rXQZfYuU|nxP?~8V&$VD1ItG&Xt^Lt&R&Yf&0~Vy3FCBcNT6Bv##W&
zNI7>zHesRn{8ips&D_%LZk-a>TcChPX<B})Gr3NL^avOZd1%5tm?~cCIyR(Zy=JUo
zR5IqwH`sS<%2%o5E0&OwI86_)sdI}x?Qa13rBCBqMeoIgZdo#pqQW#dH+Xn>j6$?`
zjCr>enQs{feE1dVeKK3OMZl4JYJ9$HEVJw@OZ+Z|!=$C@Q1|GPLaeC#Qh%E=bE^=T
zZ1~)!_j(}K-zZIk#gugE87bhPmdGSbSENrfp*`jfz{D=|%_1oPi7^!42grmdK~Rnr
zGosb5-V0a6UohitQM4i#x$f89_EVv?G2V;?Q`QM^h=}CK2*P>lOyZ~*`UHWlu1+8R
z)%Tt1-=`v-|0!>{lgrpKi6(^iz9ugWycZg(%;v)q;?v<KA$ReHw}>V?d23p`-t+Fy
zSyZu9J;;q4>awg|q$J-+R)J;6YNIHqYpwQO4<I{QFE5tn)%q4Pc60H|rJw~>S{3P4
zl7fj&L0H6TR9#Y1QVo7aSD1i5CWuBUN$<gfSql_$IFt03=hnBh8?pgrFFZz#M$Hi8
zMmkT6u8R!X3GjX|18E24=@l+m#T#vhUO}{-dotX`V9>>=TlQUc7=ZI*<8<u?uk^Hk
zWUnO};zCjv6fzHS?pPbJ9q}EBBW^C)H%bj-3{N<B+|(99iHbN9@w#HP;JM|%Z?vF`
z9u$dErX9R+i{O?`dSo_9aPXBgrgOL0jOdxfMNb?z);D<K!Tr0tR0)eR;71Lt+D`48
z`#kSQ{?30$35Q5l!EO1Hvb&Wf0#6wKHenkOa%N(pb2oyONx_dnAeOUJh=yA@!BVjM
zZ93+C?(rR3^n<fs!!ah%j;W|^5ua-3mAKFkBy@UmFT37~-XUe4+ZI-NDo-~^@J}w#
zVh`vwQ>`<v!#HUkx_;;es|it;m-_04#W-7T_k^b@*yW7;4Ai#XSSpUvrH=){CX#x&
z3kkz}+z_i=$wRQe8U<7(OX*iTIiao?7mO{6GjLBN=3SRX<7JH&``W`6X5xlly44yf
z>EAyuUvjfr(Ey28D}u#+h6&?u*Na%bj?gY`gIMs^YgBXJjxH4gkC^q+^6vh^w+XHq
zt1)9l9QDH3H0h1Cg(|V(+U0qWpk=nypTJ@1NESumK&?Pq7=ISx&ZRcs5We&wAq%%|
zR=%;RS6<fEQ!-bnnR*sUET4D7^I9w1(FbXi0YCT&eMI~P5TdsB7dI9yCAcDYq)#{v
ztYLhO&3JXJyP=bsxS+V+r+uY;UVUO(-t)oiI9ske2JKI>tJCKWq&L#NXM`#2!#nLi
z#L+73VbuQkM6;SL3L?{45^Qj8aaMmLAe5K&7elKNdXqHLB7ZEgc)*_uYa(;9NC+P;
zCeWM+X|l`pEG0vQn*yP8^9<}Ay)-YF8aMBt6CDhyEg==H);AML|Kg@N1N#z$evJQP
zw(vY{m^yxact~fX67uTwPV~<&_JzmcJ-K3Wz~A%ufn{Vi*g2o9@=_~N;;nI-jwgy(
z8q_zdzzygN!(AvMZ=of;R;E5eFm@=hqL&a`z80!Aj=-xEftV!3kM}j7>^5Z=5I|7u
zO}GqeWVF^HEOeGa)lnRixfQMiq*oMhPY&(#MrJpzs@-VoD7f|QP`jplz$Ncf5g=N9
zB^S#t<-2iB|CxzVZe4=WSkK{-)4@7T3aSE*6vk^E9rUm?nmt|z$oFH$Jn1^TQyNNd
zND^xJP5fE#S(I$(B>K$Xx?A)YaK(u2T?xEccO^PR4HDRlhPO_;W@%9M{r2I2_6yL%
zeoqu&7PujixVhl6e|>vAk_7-sK9W<w<Y~i13P|S5X<}FfEEZ!RzLwO7U~hcLBJ%QK
z%(NuFCqMh4TllsxH}WPB?tTH=5Ds&eGbzv*fvm5WM+J5kPRfynfRUMIoLk5?WW6?;
zc^ijIwlu66oandzV(6(&eVkNq&@w#x;ya`VD5(IZ%(%|6l8_`77$$?MxuZ2~(wm+-
zd`(i5!t>)i)GbSJfh8m>Ubxk3_C>XtMlZesj48)VCO$A6!j^NGv7mnBRpc>c-VLt)
zI+AOg*j7p%*_%<#K<>=dT;r$^6+6E$oY82bUzRnnmuvW(NIbVd#yakIpv-+JeS*ez
zC1Lrs9Z^igj`a8q`MRbR@PZQO+{ZXhIaez;Az3o6*9py6oJ~LeF^NrH0mdvZ^s@*J
z*)1nhr8F9f{JtK|i0xo2oCOKDnEJik{k^LL8FAf<>zgEJf}3~7b5FXKnx?{&;>fs+
zY^FW=FN=#YQug?UD$&|GGR>0mNflu&tRNEntqTYRv^pnhx!FFfna=Ky+z<Yv7q~xI
zbF2hq-K@$|o%bt<7MTUcOU=ZY4{poio2^kld5dm|-@DE=DEr0W=A`UIX^ZSgVRtF-
z=K@}IZz19V9y!2nW5vvi7IDsMd$m#7QDp>?8?y#-!c0Wc%M$ruup-Os2<*YuftuO=
zYc|8Sc->(*ZtLDzeY{@9A46#64O4+>=`~U0tK<<S^+tJvv}yry^1D<!ro7h$B@m`E
zg&Y42{CK{`m{Pd$#@6Br_2|hAoH`b!M#<qrcu>3ZZ2ShevYEa40mKAHAA13e+yj!N
zAV@=gN?(T`ZHaIMtHkxrA=uG<I8)1&e)w(!XJa(AL$XvoS}>y&s&!$*dzElh{^$y)
z<4reUSl+$XdiCh>k=>sNVIt`_d(=OD7t~T8{KbIMTrrUrY%v-4|5dNcIr4q+r>|B|
zJxLRj7BrFtp-q!KJJMc{s%}6_$Xg<n(Og^>DlrQZB+%B$ITIEYHhI}@1zG^zRm`Fz
zWJMM|A=RvwhRH=vzOw2{ky)pu(>PhB%owTPd>t=Icf;wZdzRY5v^A0JYZ4}+?%<n%
zfe$WY?~3`?64ro-z@7^<nk0l!b46)7r05nAK>f~ICSDIM?|8$Zb|H!yFSgaZu~mNR
zduO)?jL!JAwM9UMHl4cxR5hStBrY~8^#vZ)qUwRyj>(P%%G)sombBzu?t(p~J6tMj
z(=g&}ZQ<%@(LnTC1_XZ{;)qa|CH|uUQ-olLREVh*F$Dc%a#rL?qpL24#Q^wRnQ+J%
z?>@0X&rsCLw}vBd%RIB)uls<X!J!&f0Aq;6yhXj8{&+`4T9ISL58lRF$(qVq9<Zwb
zyE!sU6sa}%qQz_npb(7$5$JNKhzm2<S9FSGIf{Q^9CJr_wSG(RaBD@I-@1GXwf*{Z
z5b`(8ifv-cWctL2;4{8tUmW$LzPBK#Th4Cn>)ErA^oAQuk%~;jO}E<+?t819U2mXL
zO?I;l{gjcn5sNTU0mI6<cWb#DcM=O~tu_(O{V+udvW)A*)6t^xlG>YMe~}>&)~7&W
zp$Lh3azq%I`!TZ>C<4OO+a~8O(UF^4SsK>?x1HX3?@r@%tG`OSR7#rXULr^q6X&)j
zE7D^~P2tt^dl5oy1*nHX&c)G^($U3ai`2#*CV;)Lw?ASY+`CfHEwrLNbE)Gv_hZ1q
z^%<UvIoB+y|Bf7l#&G*o41xcG<vDcadBl-lY&11qzk#5If7)4t$$p+1JCOAS_kl|%
zqn7f*<ysc!I3u2j&3e<!@=16+H;7yZt=W=v=Pc{p)B+`hMf&H@olR4E<8MV1uoo>%
zT>dPG0b}Lr?;J1UmZn}{_mtUN%u4}sUb@b=LfW!@OfjpKt+(%hVCj!IyGN}yP@91@
zjw|^+m+Th(7-zW=Ehb+a5_vB=?WFl^&;EQ+Tz3>eRPA?~j?3(RIyS_3tkRtWx6(eR
z1sAGus;Ncx-<isPzS4V712^ub0)qumKN&BkN7oOLs*Buc>nZ|c`nZ%t*}d+8DHJOD
z4Jtyuoh}t5G%~H%dY^mggM`=61V{n^-ZOPXKtfnTlr1;u<|S8r;K^`D_)bwF#YB46
zXPtboT20Mlh<<&zN!w_q@}FO?_V<ngBUYS98$2FNWO9+O4+{5<F5OE*dCgpU3=Hu=
zyUi<p!wIAq_6-b}8+)IYcy4!z?;T*xRyH>Fha@$IFQukT>y<^na1lem2N>-&oA08(
z&&-(aF9IkIe$3GDozx+y`iLl#5Y=>yuuUF1=dOfIk|35=S0KiT?akIJef^{~<v&YL
zR&>JkAKr{+s}(AOxgT1eq%p=3D_P)EAPHFSM@98542nybugQ37@8>o#iOGh_xabW{
z6@+JX6)%{XGO<nsll0$~P=YpXzjo$?b#LDF0C6#)HV~&|@^@scWPyQ!CljW8onQc8
zUQe9{>H6h~V7X4cUKJGuqdwj=$l_gq%O#z&oG1X^U7DrTWYnA7m66^FTXn{tXC8B;
zOs|vqxo-5w#dnqA5U;h%*<x~XJ;Fae$m#^h9|F41p=&m&Vzv3c>r^G1_7+p{kX~!F
z;ZNf5jn&ncky@aA<MT;xw{Ifbge2u2w~KOQrA`K{w3IBA&AB2O=IOSXY*{VJTQJ#>
z!r)xhdV968&V@PGg<@;hSFlUvT(}5;QI_>PeAmG=<y6A+@+rTq^se~M2u^pWze-7=
zB{DEsfX#${RHIP_>0Jy}z{+QJ{1{_ie*hix6@t%-w%C4c_)o0pMf$^+d!)c1;8Jy#
z)5kyVzPvQr|I~dacWm_7H06eLZ0k|~sxIU01SP0C!A&nr3OJ*RLDu;QkUp*9L~`Ox
zftbvjfhhJL5<fon>lP9B=ooui7<(ONUwJXd#h!%5v<eA5fh`FV3ci4?Ku;uEaS~jx
zf=gvqAQ!krXoC)YU?=5YwllojzC(CY;k-}&*jgu_w4kr=23j-E{5J3T%wfP@_tE@b
z!>>mmA8w$iSopm~^NTgzD#{JjJjM*plq*9Q3j=l$_T%N#rY|67t*FaG{w|^6G387&
z+9%rRPA9dk7rY&PEv#W%RWE@MqAd)nJ|megR4{d9mNV_!@m>w`4y^l8K;sFs_;(<U
zb3cR9<BwR%hpagJ^AKPQnAgd9kDa|dj4-S<2CT21)jl9u93QnP3Ac6Gz5|W|m-2cY
z&=6db42H#(D}y1QB^chlY#=G5EQ~0yUT3YQh4qyOsk~gTj&PmEi4hj{LD9GIIN}4_
z$Li|wp@O**=z6&F^~NQjdNaKFO+p^SCeTQ`pl~qDG+(krQ^z2E!zDRm*cpCP7BvjV
zn7-LO3*CXsWjOxRXzK3W&vT63D=Bq8G`;b?E3vdUEWL48A&i}uku7ZxW8JW;bm2DQ
z*RlOb(R?<9&N#3;SV7@{ulyIL71fOk7ZAVqxcF)K7a)Wcv!7TE7FD{F%MsGsj)i89
zn0SGx_tdmH|C|(+Keg%qTb|*RcQ<c?f|y4s!Np8D9{6-jjSZ^JaA>r*mDlzSR#%w|
zQhiNU8lwwAF6~%lZbfEmX8~5j!KgIIUgJ?tX6F0pCjU*4UUvrKgDpTV>kEYlb<|=T
zc9Ff08y*3FW11EC+GWcF2>JPIr%O~G12-O@VdcQBd+RbJc#0F)3mk|fv%+ce1%=m2
zFK$_VGoyf!Vq+ZppP?7ByNatppU}}&dXjA8ZA9bF%io4AQ4BTvn9DsG7vEDBuG+aa
z1#UC2_bheO3&<yy0#Lc|c$QMf6k>bwNm!2tLeh;4-C#df+Lf;d<%F3*rTi_qrq@uv
zw%Xhc%<@<--jr@&lM>J`UCpNmq@9P#fKA_Qojo)$+9Vw(<NPZ)`=5q^-LG2#T+D}o
zy{o_`hT}5I);%=)@SI7{lYT{q?D!~s_4z#}3h&4;L|G~xg6`NB0Lua0IyX0`z6iW@
z#`}IOvc8^<Qt>Zy-7un_h4v_x-!wvFUfc|WWL9_A1bl!`gpKQiL9vS#*}HILxv1&&
z+ZS5rnUFh51oM&$5-%5DrxEpNZEZ0gAm)@wNT&B61gJJ%5R{Yd><GOw-tVNR7aUX_
z9TyijT-o6`R{XwWLa(loEKlEL#mqZ?Zaa}u<V9A^G#$lEk5saOcpevV;U#Rm;mim&
zWrcdH56{6+C~Gkp;@hekw^beVqS|_hf1iPLjV`zdSR9~QSnh9Z@s616b_#ig|6RI=
z;dWcNCs9NnZuNNesoSX?qiILM1k*)wMxK>ltW!BkS3h249W6^08#zaKeQTPMc0aa#
zV%2;*Eiz*Mz<N=$SmcN2!gkuaf#vy5x0t?>Sc&0xA&ClQ5N~aZ0}+Nk!xfd|H(+>V
z&<S0x7-Y~Hjwq(|3A=2rt}#e<i|W+%e0{P)d<NN@1D}aXOZibjEeBNW)MpZ$1R8gm
z6(Asl)hhS`%<YNQ@GDw&GLl0l<kRNv@m!M5%)XxP9q18qjJKgcIm}%BUMd?EVbrDR
z=;rO-rA^txO@eHaz^g-1!e31c=#ew+FCT&|n8)yYPI=}ob~Jt`5htl0+Ra0nX=oLT
z@{MxioVszgae=R)U?O&O<;dpfGUsr@Yqf{WJjkMV6Eo$0HErztyYa7hk)QliG>$tH
zLKcP>28`7eL1z4gkhS{qXe?d#9g*p3xJEC#-wPHE&8SZ=_sE9-2?}@FX~C~L%<2n}
zI;$ZgkZP`aH~gfAMSSd%X0-xr^Zi99LC9y|r6VT9yNNM>xs1-ItPNLYvsR;^h^)WG
zjbO@vB%m}3X@3BBj%OLFaY2Y6fpk9V`1L+<MJr|0EK5Ydanv^J#VtuF)6As?Wc?dF
zTG}NL|G2b*E?PPEYpfnN%L{5@Lp#<YucQNVhP&hGfgL&{8tO0frhMY17`BC?hn9fF
z(NP^5phTNTkB?8#(oUm0H%r6r6=_ayu>~HhfIUrMF#psB{MWS0eEu))7Z87J1?S-~
z8SYOlN)ZYcDG)6FC5_&r(tDEymOF=6@Hf>XrmMp>r5JAX0>QaPn)Bn0`l;b6d!iH(
zQbzh^X!kvoDBKXr`HaM$nl;58%7yP?B;_$!Qtdb&VxNaBm!@Z))(Z)`K65?kLO==M
zp%+aFgcRpGQTH8Q<nrQrMK>#M81&KSdZX)YsCwFdeDh>=s@j8ZJ@oWpKR&)LZ<&_N
z*D&W+&;E{ly<yw=Z{Y(C2J#1}oF-I3zw(tqvzoWQ!*Ye5>twop%z(HAVsB>5U$7;N
zBQ*fU4*$v;!>R~3DEh?(=YrEQCC?EPz+#$Lay8m-%h)_eb8$~^K;9u*DPsg-=tmu_
z5of^sT5`TOd6sm#aDI}uR(BGAZh8)i)nAIwhhc{Q`6oCLfqMT$-lQoL;0DP3nVlYZ
zU|ZN}X&FltjFoN{8bw25bK75T|57KnZ@mIsD^M53Pk4n?J(@0@maSM2r0pdTl{Pyt
zFo1#}jM5=Vkk9B&ocOgyLC+Q46$RBAu_6}u0W4jw5@a?&ruH8hp{X5kxL}yMX}5=C
zxH^S@EQAMm*QF#RCZ?`i``(w;Vlkk2<2Bntij%I(St<!35F>>$=%#IeR#@dc2MeiT
zklkkZz4^+^`Hom7)5wD0gb-@(NrBu*e=cs*WdK!N!=V=efo(Wa!L3`_N`@&XR|Ed{
zw<}^Qa0U;r^f~jV!;Oswoju*cV$yGO3F#Q*wt<N%S7I_!#1MLYq&JW>5`-eG4SamC
z{;mBs0_4~igS7kdoI3}If;eu|1)vAX1^m^frIEHE^l&@b9D0Q9ljC?n19^TiUOxc1
z$rLIfT5p?MsNdg$%4$Glt%LvS8v(BK?=^56(JM7qD7?C`fa3NTMTjlZx^G|Il%0s#
zHg%(>ntKdRdb#eG4bsJ+nE_#Ky_J<x`xkEIuz!^?x^QU_a%*+YnQ3;?9eY??N31dY
z@mds>!`Jfi;Y%7FY_E%EC5^H$A5DVv2_F7??G?Y10+}HepF><46t=O$zktl-d7nq8
zOJuir{*|{$vCnRFQll!evoYz3W*chZ(2L%{N;~;8g)|J}Y-y`8c+%`H;-ecP?*_~}
z##K1;#Ndw~TQ%WfZT_zTjz9?}NzJP}rO0I+JY#z|0lYQv5L(xjGf3dp)fk@td<9lG
zjXI&&)DlRG*-Q1ZUR^Bx&?B|>%sL2UDe39ErfoRNBBZ3^q}^BwSO*HD$SyK?YY@Qc
zS^rrTI|G1rkYJq}^r~3I4QTc3e~3(s0g{RbLmH_kw^9b8H8JP=!}e~KGLTTQD!(f%
zU2PJ;muNhf|BX@re+wba))deK3i=UH&}(bmKhUw5AgmSeV^6w@qYW~{WI))xIWD3}
z=*>#Q{;#O65EL%a8mDtE#m$tzJy0CRvU9NJY*p>-?^k``x@hI*R>23NEoEff(k6XZ
z+5_(}B%Pe3o*1#dWZL-aDgqz<JL$izj;E(*?e!}eS4~M8FQ;0WKHniF)&E(1uj=&c
z2&l`SFIcCImb8ZN6;yOdxgX1D+|z!lIGE$X2(ep*djZboxhJ4}o$ND`oBjj{KbdO!
zhV|h>vonT1GS-Mn`3evz8E+IcKue$@c1a41R6RlRp9}CROm-zZO=l;hrEReO><@5y
z`ZQ~-(GP)3T{|1zwOw0wj!M)J5!jP;JFw!#zYFo0Z!|r&>U@4|?p|W6UpTzwR3tO`
zuzrOC?+x@IQ#ta;ZWHnvxWlcbCsItz*eKk5H@clO7>44l+ZX0Uo4*4i`53P1g<72U
zFiN`|jAkT9&N_I)5U_mMBbXx0xZ)Rj8f1Ro=8il8A(||?52z1pD_V0N95LRas%J}i
zrFxqE#gaG@`E^(EJAuh{+N03|0@7Vw47YrZ#a1#^oWT!sutWe@{W_3s)oPc1#*?9r
z#A`VoCR?-!8=@)eBjln77IQSfodBsn15+V56QrFM3O(Hp^B8M(>-i`0i|HeiR_#1P
zad~PrEtcSl|ExSB92XH1aG_+~<hYE!iWIGuARcM4h6#n!Uws!oDF$tpAppe7`_zju
zy3ubG_!}8bMrd>~IpeE0FRtE=DDPNb1fTOkgcOn%iCEUmQSE1Z!t#WI(D-$d?Kd|`
zz@4zZa%CI%byBaITRt|q_nwc3zW{h1Y=GF;-F247-Fus7O!JW)2berwmG@miTe$(@
z8@aWz&zz`~qZp@RLYM}sXDbDShi?o8RHN)1BOPov$>!hADs;MIK7#oV)^JjYkXv;|
ziVY0@oocX`IjU0$QSEOy_}><hT{p_4tG`A5uLK7t!4n?P2S5H!$+!()WI%$erEuZf
z@%Bzi>t1VU*;*C-{#r|P%doZK;E#W1Q$y7mW%kNTnwUh(^i{c}UwzQiGAi9LN`<h-
zBvBTEzPpgjw+O^*d+StR0;7|285k9eupy!v<>|;~4<f>G?S}N*+4eg%EaV%wlKz_(
z-Gypan64z3B7+|T{08e!h!@6%<~i!FBSVKewbUID^W6g!*rmB&o3F*s=|A&|<KyWl
z!5Q1^e_nRCY`Q#F{Pg3OJYAZqTFHJ2dEW@S!h~LLDSJAZJhsW;jW*{njhkWUl4vM*
z#fQZ?C8c&zHsfJDFOU&l>x38z_N8f^mkc&@AnVwac4TP2?~*iG?vQD0zt!o3VH&{Y
zSN^>=;8NkVintxXcbnjJkj5FHq7y{`b=B>OB8z=t;8LX>B;}Tq0Okjv*z!F7S&`)O
z=~toqE3)wF;a<+0Yws?8XbrO!TN=@QWLGy{+$TF;G^)+_m}Z5bMOY2bj=Y)>tIbZp
zgpE$3#6~1tYgmo1JIOvT#Um`lq~U@HucSZ{nY~X6<7ZI}{spMUe3*#$*>;+{Tg06<
zMCyA;`bOP(?W2be%V}CG@(e0W>&{O%PB(C%lE_dU^Ngut((tfRFgbK~+1Gub8tS~i
z^Vn<ENhc;=PfH;+pZIOm*=fqqhsz6@{lDD%53r2JM3eN>(dhD9W_Ki|JUifC4r&vb
z+sdoWP|SI517KMNn{CpC^MO$pcju%qQwFg!&qV)Q(!v}OP?LQ%*%8N0Fh=(eh{ZtE
z)R7}+*pg0Sj=8WF<vGq`?x4vsbJ;!mNL2=KZcP{{lkcL83=CHPSj+|ET3w!abfZXZ
z%{h~GUb<pp(CV*W9TCZlGj8+%G^Xdzp6RGM8xQmWZT?hbglyTmZX)PBr5#UG4LygS
zFxGgAQLI4U<D1f4<TOgg$u@xrwo>AdXEv1#L;n`717{sXSDUHfVvvp2Uv>M4j~~!H
zfk-4r?$cc8ry<z*HucE2SNZN&l~ychzrF~Hk15S`nm*IgN{wy(0Wwtp5yhov6qyn~
zOAvM0URO-<l_8vbh#h6MrJ~$p-{8K=-DBVEMAsYqRH4G_BrZ96(pQ*)9`<(j#m_@S
z%3a34dV4T#r2~V{>%@Lf-8J%mezY$fbM6}aF><W9=@Blri`a`St2!bwKx7-3x$p0z
z`#;v(5GZmc%~x-COEZ~<>+bPvPMHvr3G}UvmAQ|gw89rK4%mFGxSu7Y>FI)FH3n$m
zd$a~~&AU{aH3%M2{OcsajhK_boqV67T0G5pe`Gd}Yz5?VvNp!GB3O)x+_=YN1g&tz
zvZwn{=jP!@AI;!0I4DMGiU_SX+dQrUBN97B(b<IiMn#!PThBu4arDo2EY-IrO8eUp
z!52c5;S(9Tr*{rcL6|FU1|;AbyD-c^Zt`e_;Xg%?q8g4F$f5AvW@D+WWsl!DlR{`q
z{o<mriW6&$dh}zdSMS$|Lwte{FpCn2n$ISb2iWJ{Zmu1K^Fd-QYs3-K#Gc9zIe6)X
zVw?(=$lq-N+st;TKyAU~>ALG4Za!!C3y3Z4*^-}s;rmYkglPxcrY!cwlvR>R@q<SE
z9cPoxS!`}09P`0bNvUw_Y%8~^G7VsAcSs|kH%E9wdtagNqX!E<UZAgJN!HfZuAB<y
z4%z2oJ(x2o=D22WN&4Rt9)f)eI@tmC@ShX?yliGRYv0SD!b;i#ast|3s~aN(`8}Y1
zn}jiuYbdUQDr4><7t?zHqfM%A9Zf^MVgke<FKnps@Z{;w$q}wJ4^teu9!m+^1vBs3
zs)|3=Gj5!V5&`d_yTSr|Yv1_iFS*4f%5DURFR~L$f1qBNA3pIjP>qaDv)$Z80INU>
zQ0rgn!*PPZgvwLdQSAPVXf+bNi09M1LHu_urA#B2brd`Sw;iof9^qaB#r0r+zlh7r
z|6DQ%7Pq2|VVB|S`n9ky`sUKDxTKV&j9=Am_Ipds&thU?qwIiPJJ;0fU1(16r5Vl@
z?^Fqg#!L4G&&=ZwaW!)~SWQmanuEZ#%1Q)P>d-Lz$E>X8G`2UtK!Ex$wd$6@yQ_3P
zh5vabz`dDR45G7PEBu4NF@zB!hAjin)ZoN~osyE0ndB`VzbU}cS`Gn1UYpw3xdmOi
zb4`g!c_WuNkjuVTboE<6TvRF;q<?RuO9#o$;OknoF-IQ+D|0cWW9d2Zn9^;Mf%Kv3
z4N*hP=3%=%7017|DFoZ}^j=6h<PIbSqLJo(R&kntzFW6yd^T^qigWeLogemS1BC~Z
zD@D|OYO%{a9i4@^<J)^zSpPv`ukgfFU+~ierW<ADS-pGTI&2JUhp%0mKdeme=YYgu
zA~CN|7%9A?XD+Fo-A2oSxC&g+sWb#&=T@L#a>a{DnQCV?ju_fJLLw|`G|(Q+ue#fv
zBKrFJre~}GPuL~s--QX7owEZZG%5Xe5V$pRs~03R-{9unWGwe}rgn=|MrNa??`Dx5
z$it1aW2lDT?cP}B8vAT}9P&*&A!&EZvsqr_ECl#E0v`$RmoVGvTQl1k+2ZnIzwf*R
z0RJnd<_{i3+dtmfS$aU-2je6*S+G}}aFfHtp?sKt@qpi;eCkklh;o|3A^K!Mu45`O
zO!uw~JHiVDuz1;^`$0DDFi^+*CcVG=QE1kRkaJEXjdgT#YL!j-pF5fHx3S#go)09T
z9DRTuG62+0kcrR*RQ|u<RE=Co15>wJk3t|>s7Vfd0s)up+f=iNWQ_@Td!i(|`33Vs
zPG!W?Z+XTFt5NFFsJY!AGamZdk4@?0`HPwWyYo0nXXL2E@kV>6bJR_kW<0d#qYiJO
z4<@)w=pvW(%OX8kPiJT6X)-1rlQ4&`I0>*ktW2!2QXOo-=Txb1+FTstkxcFoQz9ox
zsJI9^Sq~jIG4;mG7*RzmlRDvmUM=-L<-!y7`#|(}X*2Krzo}~$wuZp)H0q7Z#I}NP
zR<;Nv!I7URR#`cn3m^TCh|0LI8>hlcF{}7@1v}=uM?L1d05B|*$Z+2P6;51YlodIy
zkmXQK3bh2TiSdJ*7tmE~imW`WK6apz`gR?$XvGdqE$@YBRHyPfD1*xF9bV6Om3>i#
z8`M;KkDdiT;^Qs4^(~+@C#&~c59y=%()?E+YpNWwf!TJvM&XKMF`p*|{dM(gUdq)%
zU`=S=lPa_A%hTcs<-<z6@7Jq&RWezRNnV}mw(oJR&&gnD<V@{gPkj*Q2r$^|#E*4#
z7jILy&L?9&WA+9(mIYXMS9$7yyGBl<NJ`@08Z*Gf_<sj}e_b|DRSDEXfATs#HBqCx
zfLMSi;ZkI9;YMT(5>xM*ZW56253>Y2E3{XEL87$BL4onAxsM<&kZ3~_Rvsr{oJZzN
zM-mzK80YriC)Wk3$h+iZ^s(_NQ&@^`B41g!#?J5E98U&jMuU3<fwIz-+w^R76m%M1
zcB?uHDRWPi$Jt3zQ)b&WBoIdL5rJf+LP7iOs&}<hNHr~C#)lKhSMnc_wg6&~t!1O{
zanPR2bYp-lF$WmD`DjpG*rmpQU7Fhd0HnCPx|WTRl9<0xtmphO_Yre>KtGwhpssNC
z<DVoj0*_JSmqe-5AQbJtQDC?)wwD1!fmwj>6>=LjRp<hL*McV57%S^hnuaAWPOVSW
ztiZWbQ7FvW9G9Gf!*yB}0(C@Y;qu+s?7^0^jr^Kn(X)V@#dc$L-eeIk%FN*OVw0&3
zOBR7=!p-sS*q=r=jZ~>bsI+SKRIdNHkpBO<H25A;9=H<WL-<*k(&}0KsN^Mp1@4KO
z(bl(6HTZ5K^)cxPUTVtB$PT6%M~pGQ<3GR{=4RKS!!bcV!@g^mfuPJja76rghNW{B
zm1XuOD*TOf`?GdbOO@pEE5gD;gl9NS`%n$s7&YvK%fBm0@Xp0lbS7U8;zIj|TLXYP
zsU;xltxRvK#_@uKFDfRbwC;Fanmp7pe$Du7E54RiT^KUGi77HM%R==$WW9g2(J!ZL
zl;#eSboB5bJW7{=N#i+q?!F6U=;hQ6^Wy%5I1b)mSKr`+xnUc|o<XoH_O&kTAHcQ1
zT*h^d1BT0mVz()@E`ISXrFQqVNdL;YNcDuY9Qgc(!S>3-Fdm1P0EXcOH^2_|zJ}x=
zPKVzm&{&$1;X^*e<X5o*<#P_$)Pd(;6okv0h~%t?<h;W%=}YS19h#O7*LVd4_C>)x
zgys3|tW-`dC>E1$E|wRZu>|ir0ULEQ`KJ2!_%?6Rg6#oDbzYu727VRfv8CO}eZYf!
z_!`EY5dv~%g+Z+2C*0qiDaHHyfACm2*qDb>FRHD13x7Z;5H}H5HO*oxK%BxpfgzA1
z8LJ@_n$|Z-NlEWaGBRym;JC?5?fl;(oz-g*04x%_w%kR&B~vKqk!J$o@h#298GtKW
z1NQb(0QsU;z_tLCS|y{TV{_UD;HZ!>%O($8>7FNjm-s(l;YqK;homJ7hF>v|4jopK
zGcYi7*;&*w*7pIxlY_Yo3Lb9|zbOrn?gG^z@rxk+r)0C_zefPnDM7MijQIZ)civx3
zZQUBCHv`hEv{2$l4<#T5gwTR?uz*N}fb^pDsx;|}fb=3wrHMg`N<ahx8UaC?fOG^j
zhN=_+;aeQfIrshpcifEpkg>-~#>m=xwRg_>J}4J!D~hxqAlSJ2>-fZr5Bc$U?Nq;*
z!v?5cY)VRE^13Ac?;0&iIOSp*5=}+r;^$X?;vrZ)Z>#}r$|(cdG~<#1C76w#khof8
z9=sAR`@g<3dHq9TakwBt48_p({RhY=el3Q4wjlAo)~bfB*>~~$4sRyKzZw<TA5S`i
z_LQ*$pcHlnUc{UcIS>?Uj?}TC=Un^C@$zpS^|)QqKS?8h^)O`;DDe7MWkvP6{q1h}
z_oXG}KNZLS+zmheKc+8#_esK;yMG*L{`viaW2N|ieh+K<S0Ml2KlXplt&>~{NvT_g
zQbHh*v3i&On#RTvfO|XM&J7L?$(}nmbBvq=0Lg}ymX^9nR5=USqhVYs@q&Vaqcbx;
z08^Khmgb<Rr-umws)f(^_&CC+Xvqwa&0gr+P7V%YK}G{66=+(>#igZti|kO9H>O`#
zR<@dW_M?`E9VRfI^b;5|{fJ!}u&v$R+6rlX)Vla<eI56?b*y;|GCUI<O=V<cBp@w8
zAV(G&*&Wj&0$+ecx0UFPfx~nj+nkoM?wFXEX-TW>&R~|XoMt%?O?hp35-=`X3uVFh
z+&e2ol3P$E$psWDj&(M)+1S`z0{l7PHK18rcE|+h7`TO97%Z|h;E~68N=Qif)Wo!R
zySlorZ=6EFuUB>s0pINCSw-yb2K6z27uXOj^hcT2<8T9Dq4LgG_1V@>jzcG0i2Be7
z2zKASOib>oTKbm`yOHF(r>ibjAM429pQ)Rsy&rJ1jF-o>xB}&5=M%t&c^|_;Z@dvB
zcxv0NtgbeJ_+IGeeUOcDB#<B7uA)B=4AQ*-ZFG9u+1dHLG5de0V>n1;u#S>}04+C5
z&-OqY5x|Xi=t|<s%E%CFNB1%{)C1meO5oyqE+2!L+eZ;EPIT*FbbkX)HFU}u$)jsN
z;Y9k;9PquSs_J}(6|~a+gt;)jxtR<GOyIXc2_CiMsz$(gSluC0Zd#hLrQl#V=?eS?
z^fW@jOEGcTJws2iWaZHzZ3G#dCA{ArR~n51XR6YJ@5N+L`o99Az8E3OpM^owb>MK2
zvY4a($6d%cMg+RCfO&bNeZ>2gwX^$f-M8RM3jgAkd+AT3QqZ(tmHGU|B8$!jeV|wL
ze~`>A8+v9472&+imL+2YEi)@uZg|Uk8)jOtZa`Ao%B4JmIfW``8Xh?&^Z`%XR^}z2
z6eRNN`pc8ky<;a5&NsC3H-KBxOq`{$04$#7<f-JlpNEDV0Wk0ojqmaBrC?QJj=2#l
z>S;~-29S8IPig7q#%LGXC6Bi%g_;o)-_;fnB*en2?d?FGTs6687SHaWxs9dZbUGB+
zlx{IEse2cPXM$y;)DyKrS0+@R;lHx0Il#TkilC1zgF*`1>XckF9vLJ!MI|KOO-95;
z_Lu)><tvcX?Hq$MinJmg%RZI~$UYd3ApP{WF!qSHpG6rEWta<L=j6_By^h!qZ;G?g
zk)opB90*}blTzqJcj_7#J$V_n<NXX@4kXu?B?eEDwa)+U+Sf_AAl}PA-yRi703{kw
z2pT6)QJtX2pr@N`uS5n02A{9BsJx43GZZR1ktPKahUMTN8vtAH1y2RO9Mt;mQRHv$
zw#U#v7a-I`GDlvX1cxy@t<n$deLtn<z!55tFN`Oj*FxCGVq&52Y|zB>>pe&_FR!N>
zzz5-NIB>OUh~C=^JsUuJkgTP4Va|A~D$$%IVPWpdJZyx_3|g6349G_0Zu6~~c{+B7
zpE3)LMRUt^u}WsR%Cr51;R}Qiw9V98$|=jI^U>`XFxIqa9>GgCv>!PK#<fF43CE6P
zZMk<DLwMh<Q=fo;Hb2Q%V6N(BpX1QpR}gB%D=B&oLa1!(6@o?&Dln$G+xLF_Hoq9<
zO7JX0rs`I8%ob3xP;-;ywmU%dXbLlX2Fe}-O3qa8Kp5cy_C1!VbtiE=IqL@ErHAP`
z7ZxKt=^1yi2nM4}cp_(Mf%Omv+zo2Y#N_Z$@yinpmTiU9F+wcBK#N6V0aG1<JFy<y
zX=+~+Cz6`*t4e|h_(Q91J~kP-bpnAj0okFaimY~s@t5m?8<$MZjnXORtu#<g;-FYu
zEgC+56&(_Lq-%gOq%mPVunr!A|6u3tTcUSi7Qdwo(N^!XhgiZ-i|I;5rlBLmkx);p
z*3_f2M{;5*mbclAzTH0#kMa*2N#?H47S}5BK3U;{a%ppRymM5zTc{Nwp`5YY&^pEn
zu|8&xn?YW?W#2(d@aif>CE0U6%!%#rh#|bu6V1jVwAQte9mziiN_6C$2JgLoFUH8m
z8I?f1MMu`%cAy`URN~|6P;+Z@!}Iksra@elZ0KBu9-F*Agb5LciI*9*Qx-aQQ{c9G
zdUrm*8MY%cVRUq_*BEG$ULv*io)=Hg-A#che*B1hSi_tw*Y3p0VKUY+e4VuJq+tk)
z(H)?PVtWUH6k-)9sFkq?k^V&o!3pR*E}lxNtqjvN{Up<@+|J!Sqt?eCpi&OTgB7jr
zban~jG^r^H9`A~2pNT`zqLyrQ!VT49RJ=hDW7n{SF6e~qgp2b)Wly3q1z>1GTyb7}
z#vJawl8UjlQ?DaFb9lkGvbONzv&rL;h|G3XHdBV2hWduOjTKL%HX6$+6l*3V8Z0KN
zd-E!dxvrIHI^}PT=>0fmfpFA==3R9&HIigcdJ8CW3ItX0F9cAx{Fz9~TbX-(b)CAI
ztVD_xX;0`|nsc|!y5k<*1IB<8)?|9i((lzZ-)OYriKztpIBn*0Hx5z>KSYE!3WRxL
z(z?+J<8kUlvPhwfks9vMrR7~>kWxt(gE7RJ_x{jFFa->(4nAfNY~S;PG%<FC+l}34
ztvA65;%Ip*W-rFxAWU5IoYux_NTg_o2qa#Jtze6lpo7)?>gpEs+|<MRv5*i{oFPEz
z8IOeXHA!2T4iBla$wv>$SxI7oJDl|?x=fv^k`KUWCh;%Gl2)C%?HcMDUdUAId<|>i
zpa{r($3b=?sjj7|#j_G41KA^@Lc4zu8H!Vu2Ly1iifV=<h6=SN@#fm|G$s_vGBwrh
zsoUx=Hbw3!3jAmeNA)W62KyJ4D-5x><TIWq73~lq2x^c)u6&i2fH5z!P;x~9r77w&
zJVf+wM)iV>ceb@$-bWGFNB2Mbv^DmhJDR+;gP$4qxxH_@Kc6g5CXd(4tt#Cpu>)kd
zVmYi*?0utCl-5aKz?!tm_C7BOGP$9T6)9(62P3>6yI^@Fsg&?o)_IT-PbDeU>CtX_
zDPJa?5h1e`q!^HmOqp{Ha?6hiPF;GF9^q;JHkf6Zg6u@NIQ|Z5V=i17lkv1Iay4@0
z&6jfsMWKN)-r;4UkmK`;B^U)ikGhe%spt}uvM$8+ST?DT_s<DVRxIOCza(l&Ns{Uo
zOx!cvrKMgY21hW!Z54vM1GQcnd~sS4qnlwkZ;XKdxQp`#QP{J7ODXbOT6R3xrjpgT
zcBTw72(Gdo1bpg_75+~6Mz_JNi_0qOsHAR{&Xn`ItP2<Xa&d%nr@1F%={_#4)7JvX
zhE-ahd)U?6A8Q4PJkaFB6c^_`q{y*zir^hpM|Y=+6O23U?@;c5Bo7VFqHz@WlC!@4
zTo*J`OB|(zqUf@^*LR&aC4SK$7JZR7Tmf%lC&AX;-A&J6So1L)a3^isZnZ8mL)(Dq
z0+R?k+**6G9AlXpL>`+9P!&fh^0g={E58gn-J@~Z-@07gcymao$InJru%JF+$uU=O
z!oEAu)4ylXCHNCtsSKFYyxts&jd6Ts308uI*(fCzGgA17Ia{%!_=b%+`P)5|^^t>h
zarYfXEg4C$B7ubSHG1)|)68r_#SRGlxDqsNMYV&-2Z$=3L6MD1-8e;q#&Tk6_Gv$`
z$$5}pQOJ#zbinoHCB-r=yh?Fx0P7}?rp<>al~%N8idKZPLRkVw8k6}+HWfx2M?-of
zq9bp9tkbw)-i=VsIlE|lYvLLRYj|EIW4hhO7{^X7tk7b_;7b{>LbP6V1wnTN7|Kn0
zw1HH=WBhAAti%Q#&ss^vaR6N1HH#b*4M(ZEw!22{$=&-6uZBIy3A>*jCtX>k(Go+s
z7?vtQ3LPUqWC%la(u7k3NW4GqUrV?U-3?}il5FrK4ID3;F~0m_DLMgp-~qSO<HbgZ
z5kd)JbTjQr4{k8gO*h~fKymPwE>`h=fA9dZYvTkvBn1+7MXO@9)taaR^4{7TqzI>+
zfD3@-GBUR5L1^+ik%lu+f$H2=&Yscl$AbDakNUWwnI(r{4v4Qy?9A(<B%R$V+RoXl
ztt&w5x%%<Fg%@?=?=ecXY=LuQYwPH9XJ{tKR|m#Hsmsw`y@Sjx`oFK+Xl~l;%>Imi
zarz!*3j!p2Niy8R;aG33YcUNt4DyzE4!tV%9-<(YnU=E;+s2B4vNdo(<LGBmH!kWK
z;HablOfOh)wBkz^j3L?d;8@wGQzF-zxDY24OT1^mhP#Jw(mghZ8ApkuWlgLo8)Fz_
z*+9Yi>qK#a%8U}0v?K@gumv4-0hWC@v%M-Qk*>_}=Eg}s$j#+Z6Zy%6pnT~>m-UmX
zOi&@0JwwSnRCB(zthC`YJ$uh!dWp286!f*clI&j3$GAuHL2<u{d6faoYs_j}-rmh>
zonbd?H!k@kg#EIe$*#H6BCR>n+~42dBjRQnwbLjN=Y034)BL4=$+yi}yMR<Z*l&03
z+R>z_J1=RXX=i7r=lVWjCv59>JFKo08}y`801Hd^YMn-(O+_|w7;e^GS=Fh(m{Svl
zGW~(QxVW)#(-5gCl6vo6R;JF)Dhkdv@7cMV)a5l%YMEAc)WdV|c|Vqor>7ePhUe;*
zPZ34=n9eXyOPu;LGR5T+6WSQ_{Xg&V{JzXVeMX~i)`v{ZlJ_cZ76w*EeSB9Dq2aRA
z+MYHs#@xowyIB3JMm6EZz!&qiD(YeSxB_?E{$wkiaJUpK*(+rHgZNDuSkGI2pAFf=
zLX3<&Y+v!Fl>U<=_O|)pz2{RoMMk=3XfZ6@4`DD^4_P~W|942CUX}YCIXRfaF(XWa
z`w2%|k&;qI*hmxNs-Up2e8|N<>7t;Mw|(ZrUmu|l4uAKbJkt0;AI#ArI&ChTo_=4|
zI=$}fx2>&!TzkCcV8NBEN*OAwc+JymtE*WfTrp5;3FhR7iF0o51e}w^nKS7+A7ek)
z&FKs=<0oJCoZq=>@UbK<ExBvs+qZANWqhAraDApTKh*a<&7D)A4PCC{Agc`)^s+2h
z3M$WHM+0(YO&g`p#Y+YyUE7N6xmO|jSks(-f4WZ%qJL9AaaRQ*yS60lmN$aitPuSj
zp6Oz@nK*7D{-VmO>nmh)<V~KJMqwWnf@wF*x6VV&_MN+x(YE%A6r!tUyu%Zh?zr6C
z$nj7p+08{m1BsM*^=WOa7h5c^+VgRKVlH65rLbh22d*08&+_V1OYaR#xNR}&RI^Ps
zU+ygvGu74t0dZm-_ELXPVBq%3!9n<|+={jz(+B&XQja3DC^VyJoE~*2%}rW-&ae5{
zeXhQJd1q}V8#PeL%@u7!UHAP#A`rZ<!v-N)HNI^c0ewubv|p<QUeSB*95@0jS7p1(
z@Lp$?Wr$Z?2e{$_=RB@(hA^p{Bu<hQ)GQgKn3gAciK3NmBQJfc#~Od29pw;RwGWfM
zRB&|Npm**La}h5qw)nf_)JE>x+WRJ$;TN2<H?JNN{V<L>KY#vg81BN_VX{)YULa+J
zJw{tK$LZlxhY%GP)#sf!tI5QRG2u6*g*NbrMzsury`krk)LQ5V6%_LW=Sykfl68BV
zo6lvIs=0<4igZ|1PUN>Q%cLJB-JyU-038UTAulJFMO%^T4*Lp&4=5uQ#eJ;}f3}J*
zn5aD6rjw?Nn5i-vMb%>?<%wu?Q;Dq#C!V9<Z&z!pMB44wlv5uWXC<>YpxnUod0x%t
zL}N^-wl`OP9$UQNDc>5qQ~amm3;XytxXt}svF3u(m%3{8uO1{VR&sr@Re3hG!)>N<
zjl5LNCvI~!uh7TcmiguGaH>gdqgK^xt|yx{%UzAUpZb51b_#3H{4V&^8pLC{dHZWW
z8Oi+?Yn75THC_YNAEB$)B0UIsHQulB<dCZegAWTXtUvte*L7pNUCu1zGDPfJ(nzjS
zg7%q!`CmaUIak7;S6?{Hda`KWc4V<a^V=!?f$ICYzMcN>CK);6f<8C!YR?gLdY4>l
z0*{E^Lkk|yo#To%!@IUPZ!ON$+*Q}EF<nszv0f=YliQxIkF?21wA)%GtTu3AFBd8_
z`uH;NyQ-{SHQTuC^{m;RPR~j#xbS6@^o5k~Yr}8wzYBPMa1SHcs*gK^uIq~})*Pu$
zC=KRXo8K}cJjZ;nfVc<~W(j4x5k`|M-XaH63b%Y;wn=!yF5kYOGazo5e*DfKJv092
zlOf`%sOw+X4AV_K81nu!a2n!|mi*ITydkdm|F<d5XXcRB@Y01cMvI;I6yRm3XR2F^
HvXA~R`Lod}

literal 0
HcmV?d00001

diff --git a/docs/assets/graphs/unified_bucketing_example.png b/docs/assets/graphs/unified_bucketing_example.png
new file mode 100755
index 0000000000000000000000000000000000000000..92904dfbaae0fe27f0855c165926a06c7e80bc61
GIT binary patch
literal 24883
zcmdSBcU)8Jwly5YEhtElE=@p1sUk>+AXP=0h#*yABOoP4dPhO&V5N5iL<FSwfYKDC
z_Zkho#sHxN65eN_`<#8x-RFGY`QG>5`@Mho%P)qkwX)Xp%sJ*5W6n@rZH?o{P9B57
zV8^dsQN00!9TtYc4uw)40soWQ)At3ukU8JbP=*zBax8#1hixuvU53GmqN(=o{s!J3
zeQ@QrGYrOj2Kq;K^~UKRFxZ{XS5+_H^e|tU@ZT_SO%^3d85{PGjgyHoJzMy!6frNs
zD2={+(}6nY{awAMU+UP9Hf4uPW?KXVS#rg04mP_!5iB<L4L{?{!`VJ_DrQ>i=I>6n
zdJL?O-Yy)HIDzFpe<91qvkq?KFV6vocanTMDWp=nroNA&wnl7HEp}_+!F8#-L_h3h
z3fimS3Smsn*Ygj9*SC!6&=u=mRi*+jX)l79!Hd$Fqu>X_0<N7f1~0JJr%Rz1fvdJ;
z;N|9v?nB_^?dkvHUreKYZ(JrY;W?LFX40{=F!eTDD^W6Qrad-iWvpT-1_raHmi4&e
z$OR)r#*5p`K3we2wHthY+Y=>jJ?N-@m^RO_uyZ|i=c`TcHg0O&=kCgA*+Po<>eU)|
zY-dII(8?;(H+_3q(6HdEwISx4+@!b0Q);#s>X9d3Pwtbpa(Xf~hGI}|VOuA$b{`-A
zR+gnPxja(RM<VX_RU)T)2MjIDOpd`MG&QMUwnar&t=S&~Ddxy`Dr#qh%d#9X5#AiS
z8h7+K<~7)=y-Wub_V?Boo@;Pj9Wr(92)KX#1lsyAe0{E~P*`T1%_ix2d!M$Pa4bwo
z2HbetB3m9@KEhXU!lh5kZurApa6B%r`E*knZbf-CcRmc<h?rPce5F8W+8N_}{W{Kb
zogJy?Rc(uGx>Drf-M)mr)XuTFUQN+_;RWxWPp3aI!scngqx^AF`g;(=eP?3h^p}@?
z8w-8wJ}$u_MI6$_yyp_65azKH9<Xn=R@bRugsfW;c0JM}%B80~zCEHCZOsUo;ofeb
zFu%0fM8Gfg^=4_V`y^S@nsHuXE*$HbT5P_&kZ{Sqw~-c}RAUunJMmz(YOcIuaQ=!s
zmbisZEq54sfoQejy%Um_&TnPxa!Fg&W~{iTYjD|Se@4`Q1Vx;3PV$_MU&sg-@iJ(C
zRWNBU6uqlb+v-vHq{WeG!ZjmYW6FrL!zY!eI&yiqDAU?Ck1o0p&D@h^JKvpdH`ke5
zrP#Dmuy`28AzWGt3phLepz-*^pb4Vm?qv?$r_!uu#lK(GH+*W#QO`^BbjpobT+noR
z*RY;2;FDH^K%+KWTQ1wz7JB$|h6aA8B*&bp_2}59E_yH8nqoUEekqSs(mI#Fzg#4=
z8t$4vFRX;xoxslSQ1UL4hZ7u`9$l$z!L@Rywk~eB8KNE<Ej^#~OU1Vh=G$t~r^*?Z
zHtj~AH-~dE6motKX*d)>QU+URA|&u*N21gmlHKxasI<7>gzL}@d9(C`h``a+rpv#}
zJ7JnacQ!vi=UZQiOOR<WEwWToypnq1u5vYrur$Waj^(5IFr~nlPjL)+%Hl8sK03UM
zj$%2VMmXcO^7oBCox%>hIc{Xixn<6AZy9OvzAt&03%xfhweHS5mg$!iN=d;h9fFlW
z&8dYm&kTj&)GSS!t$0<>eS)t2NwKRSs>IU7)O)2&CN1RrB;qg^(!%1Y9_+yz5De1T
zkucAh*M_sJj8EnTbU8jK66Q0jmIaYYoD_vPMa8w}(Hw(Jypyd@Sj^+*GRLmed1LFO
zxVKwIwpv=7$-G_oY+q9fVbbK1oI7QNYnJVBBPuR-x&D)cE-nhv#O+!Sq*DhS#h#eb
zy*kW|k|M3e`Z>BW0WOBIhmB8<mu2Pal{+zw@lM)U#l}%uDvOUDf<-ZdoineVq=vP1
zcr-8j;4~ve0rG5JW?mSh83!hy7A5y_DjJ^B&iXPvv$ffXPJuq3+;TaLaK750>9TKa
ztc>$)|KM^R=9I$Y*p>@CO-s$JtZH|9;}a1mnR+bcQN^vQu7Ma7?<y6-e|OAz<XlT9
zWwTd)Z(p@*?^XVRS#v$1a4Z>&Kn3F9P17KH9vbybsw*1$-$B65@2GqDV|0Y<fyQNZ
zM6(eUJ)MMHi>f;fhxD^{*58YBGHRLvX%&Yj-=$}EnK?{Z>|t0qozT4R&kcI#D{PHB
zmsJ@uwqu+RDf#n&r5zS7s2n8yAWAnaGncRV(PrxYPU>8-K}NGisngq3r&?T#%a=2;
z)8w<1lqg&zqJMBT86(a}elw=b9Xq)x#+9L|rjdVgkz#n+cC4j;k(pV+z=bSSFtVRQ
zMU1;$x%55R&jotPx{&bT4WsbcumU<WEl$^T=@S`vWojihVK!ycN11+iI)|>6cAulc
zOr@bR^N_VaK2eobmAB`+U20Z}FU%TlX%0vfqqFHbae6kKOtzi+@G+PHe<|#aQXbr7
zqDPv{K)3YbIb&0aJdbg!XoZ+|J|5$XjNf$x%`3L4>+BZj0-rIH+liQzFB_N{E(WF$
zZm}3JVtt!J=&RH;ud8W3i;LD3EZ{u^BZhV#3h>JvAtw?7`x$iPizE$&TRGiLMch|L
zU$@_5vSL?d3FWgxqkPM%CIuC}5sE6MhpP70+SfPdm)G7lmwPSd<~+P@fSb(R)iOm%
z9tz9&?aJ5IEFH46dU|76w`j5-VWBF*>6lrvLNs{;T|T+&No?A&FeJ4l-RG!kQ`Oku
zo%Ol7ToltsyPV{h_D<vTAaT8u9Fi@{4__5ZcCD~;SfG=18aGOmGj&f@FfXipeAN(h
zF@u}*BY54%^!XcH35(g4u)Nn2VIr(WzPpx*ul1-88{?UIVBG>>RTHwPL@LvW%R4X5
zXs+Ffz*Dp%yDJ`i|FlU?5QrwHqZn`gxXbB2j(dk>bE;V>-|W)!w~nx6Z+f(PTxwj`
zWb`Zz%COMftY*Nw$kB?uwVVc5{qn3<+xmA$gtv)fL4KJM4;KqIzvM$&UKdw1vC|#(
za;`O!J+{3-bt^PZ*=LhLzQ0w!Pppbq8D_oS<kfs9i9glnp`%?QZ(L~1Db6Qi`L`Ux
zDfPlWWXH{RB*s=)Gqs!`3%GS~^O`?;Qy*ia=t9o!gJ`UkO6boSrTu{-;S1+(v8I#m
zl)no@vSoNTUK%p(n=HnJ2^M@#<x`xaQ2g;|bks5IqnJw2p%R;}&gjsKG^11ZxfSRs
zu;31eHJ3*^a*WANYEEUTS&={-`Ld2PU*xXvT%X<SPUB1V-Slt}3Li=ovr>0lDNthx
z<NX#AkVSctm*af7hC||TLGzNT*-?5{2^yD^!d@6WA8Xjh3sywTeSoPRr)4bZXAUzL
zImU{ZC8#j}%$^%2JrjHcQY#s9>XCU8Pjp~+mGc}IlNDLh<^4Gj9ykh`$b1e{s4azU
zbENL?&B-lKo=8{oL0O(^^SJI1zCE7rWOkyE{-dhHdemfll8p1(Zgk)toi$i@23Krk
zcb9RT%!MrJDkX{}T=ZOP%jWi?VZE`*XnDS6jp|$asfUb(3}b?aB6a);ePx#{=nL7&
zv}y|}MJ$OKR2mi`TkM7O*c{TkvtnlM6Rs=xTG2oTnDmPyobqoZX<<rgY8BD8YV>4R
zZ48fwu8Q}ZJfr%I45lF`44_@qAssl;MMMXn15>wY@Th$qX<!5j5Xl<!?tpAt6|u_0
zX{K<xWv+5}wPxK%><&MSW5tD*3?_YT+-JRGC3YbaCm1Fs<hD8<-x$G~!&e>M#Ci%(
zo_?sI=Nia$4;Vv>gjehLNt@4jo6@rRG@`x=3%&E%F^8Sg{DmUL{`Vo~&9Cna@_Liq
zCvVpP5KBIk78u+=OXg+0vo>P~z@J_JTfJXMF7*H~u8vpPtxbRNBwYz@u*5G8c#_0Z
zcdnm=`}w56ja7%&8pvtiOw_o~jRjKBYTG0s)n%97mt1pfe?rBQEo@Rgz4KboaQ4Nc
zM#NS0xydg39jP)uBB9h}p;sP@sPRmcbi8J<PZ(O?>(ABidw;uR%_kE8*ExR`NyKE*
zCKjC<8_6yu0xs9nZ1Cguir@5ejov6uxwVyx^%MDo35xlM80H2&K{r1u{@P)BENx2^
zXQKNygi$BgVk-J$ZlO~YynAas*L(9TCPz_ICoOYi+=8Y~WsMB-6`9i;H9YU;_Sf`k
zm&}dY_Jww}e$7A?r6g+Yb>TF6oc9AI&U<tO@qmbMr#A|^=T~F0Q$yR$l>VF26V>)G
zgzu8cLUy9lHIQ%fg!56&;Vg5>gOsKsRnsqgj$F!SN7RaG#a-YjWRNRIGJg<6)+d_=
z7>@hxt^_|hl)N)5RUc0t_`NYWXQ`m3AQ$}&;WNdqMRd~$8F*_@zYCjyVuN6jxJgoa
z&03paU#VT+nor2LnZ8t#i_<LkYTeePsoQLfiPIOkiJm03J89wQAqsAv3l*sF3mM|{
zn~DHVdy>}LMRhVYqC3}{5f9`Q`}dD%v`1tZi~*P~x3?nV7oNFfS$B)HO77Ac$E?&Y
zv8fWB{FdNXP_({2xrM6eL!JynRTlBxT=r>>8f*$>=u=}c>z`aZ<rWjB*b^zcNY1^4
z_TMM&cuoaKcOGl@O2*@a6I3+d8&Y)(TETeQs8&!;R3&fnuU^4bE$*_mF15zLC2-d1
z?vG$14DOQlc7iWWtXv@ycoJ3%ZkL=@*!$7AGA8ckHAR#{IUyLQbeiTuIR{eS6~bA}
zJ?;jNrRTH_b{N(B)im4ee%4UzHK$;2ULV+^pykB#MPsMm<rTEyPjz0G^rLOwjlkOV
zWT=&86%HEdgvpw71T>+w<lQ%X5jQRBYx4H-TP&^@Qp8DIOG~%*%v8BIE!msrAPgD3
z@8q>>(z`9iE&Fa4KUSoN8^xbYtq=t(g|Ibd#DWp)%hBn=7J9(Vh+25OWUfD{WsAQ4
zbx$2n${Yx@SDk0S=1#0Hb{X_nK}lhSVjf@pt}s2Xej&;{FR!cLaM52hgJ#0HOJOB7
zw`h{Fz7bb}#19yPVk%=hF%13G_RG*nu{tgp6hfCgQGQ8Tw}`9c5-OThX*v*Suexm;
zl3Yh8C#@MH^u@;gB$Heo@*9h##<^_Y-OZWgcZGiVJWo$)Toc2(b4#ZOk5j9JWmCxL
z3rXY$2D#6;(N=DMA9Mb;O3wBvQl~s|qi+R!-fht`EV<PKE`|1ScFZpr*b);lhi4Dt
zB3AmA*PG@9xJ=;9-p7jc%|zYYtP-2OBOQ(2b5r=@s&T~9YYy#n6m$wqW9KLBeWG;y
z)?Sun`DVUN#F@279>uX`*tzyM&S!FB&oH_Qq3Q{;Mxv-j57f5y)D9+qm&>rf2veoZ
zaf*o|3E!^bV}rcIg*jDn8}BM_4llpU5ywPvI-ePz`_=?!jJs`R(jpm_JqwQf=oW(H
zwD_7R96AKjRjnJxUdtm>y{#p*iowByGTG+}CtrC>HBndZrExz|BCM;{VWha$472-=
zR%p?Jf6@>qTcluB<93tw7#@F#F*UCQdOm`vLX0?5m-Fc+w7x_1w$_4(&`6oRp@>El
zM|``wKWWZCrTr~!QU20#PT5k)t^4$pOk>L|_1kJ9Dh|=ziD=(|dv<nYrPq<qZgMDT
zIQLm<<P9H<PF@{8>L72zBKN(BVZ<Zm9GCk_+0ZO!ev5dYS#l8--X;X&FrF1_o-n6E
zG?2}}B8c(T7>D0&<=G8uySlGa1$1_*H*33zo$w(pxo^cU4j>X+y1+**Tij!(X{a&Y
zw(=%J*X^T}ag(yU9M=6qksfdys?8zzUO#t(2JwAiB0iO8HJtnv=BrCcYx9)K6qA{(
z92HxbEOXb82Xo2rARcYPQL8Tgqoum{IbnriF_pjEX5&)y9CrS_J;KB|NIE1J@z{KS
z>n4X#SDaWTk%TE}_7%IQ>F%{|^01l7&}Cby$G|Lq6meCD<Y!gGXfwmk&4R^gzzKRN
zQHgb_fkej>6eP9oexqnv?@3I>#l=Ag5i}|~*J@9p9nY2fj=K}_Lly9|E(R7ggv6+!
zkd^*ztd@;`7rtJ2VqqCkz}GApv1MLb{SKkfJKUR5Q@`|<8v;vfK1JHN$t8<SK0IE`
zFVPrRL#z8V286R_8jB5NJ{lR8`~|a&Ty|-WvaOTe4&f}4?#^Y27VMm6luPRnpgE1H
zT5^A@G9JZQrpqBMA>h`%=pHnsltw3qo;VNgdlq`JS^MGge5=iBE^Cr7P3K9tg~QnP
z^&&nC|8FNO=Pp*82iXctn26T?P<AX>shTRYqL8BDu;Q1^9`0FPjIbB`Z0dlWbcnv!
zBx&~^?20ByYNXN$itG5TR&RmDYsB~#<#BEbX*`AE_s3KspRIi~BJo6W?rkNJDW!`v
zXE7v9n&P4w8wf*ae;;&D^(jVt;cH8iCdqej-CTd-ws&OO8xkp86eS&t8|(gx)=KBE
zKhu*egnHRhzq9QZhQ;-0aCcajL`{RIv+rj$9x!aK7ku!1#+|$JC+<fUlwajoZndH5
z>Bx{n{@`%Er05((g~!%;Bzx@R@uwKa6cx9g9%sL=Njs$HyRwXsMFo2lbPxZ=PfD3W
z>eUmB`yXw0M6BQ}_Luzgg;z}Xp1S1tB{F9e1Z@)?_v8ecoU!9o;$5wX+qTVEMI|<6
zs<atP{&E~lq>096Ip|4C;M3vD5#kznxtJ%mM>xVbv+t>)ak|U$Qeonn$m~L7a7{O>
zL{OC5UiFP-R>XLvSS%jnD!2P}A~bnP`t$K-vIptI<v0bG4L_?^ISH=ot!zufTpiar
zLrWHr>Jl`Si}iWZAC})=^-h`*JQAM=4_OlIajDL%z8%EOpY%XEcuXk9{WPj(n}(hC
zfiiJjes{T~tV$Bcm0{jOc4MQg5Pm_>gj?2YrE?M?2(d7nPMMbMZvFkW%Rlcn&Q~Jk
zi3sCZR$KiU1>&yn9<J4yEnQC|IvmMqoJ;v!ZtQHc&8moWI65FBOS$x+?fd=arCQ<&
zqRTq-ZFG8xZcPbcv<LlpY;v=F)YL>X*bB$OIgnpDvAsJy($8A$l0AQRebv3brjdZ2
zAF3z0Iu;~deR;NQpR?J1FxZCvR+D7Npd9^umu!B39juQE2k^a{M~r>FtRitaGWl(t
z_ts#{EBz!fxrH|`AMO~RbDOsc7HPQ_a<(Noxkq3sN^bSDHAjQQYCrRsq@sV)$7Ndq
za2L1KgJ^p=c@nr-<QDL;sBlMJ|5J2w&1mAP`??Ruva1`NdueDF6TX}TOXH*OZ6Z^N
zu%aJ3`#7<arrfjaV*Q!sEgU9X6r9@m^v|GN+sZrb*uPj`i1BqR*IK0QmfJN6L2qNE
zP8n0Xy69TeYL|x0+nO-mRTIaj$rVLnq=cHt++A*mqTk;$zF2zACmYlqHatdoO>>bk
z*Ku-;g=`aUVKFCqUY{<%yeGmWTp@d7)15nwKZse$x@n|*_-aInV^JTY?>#Gi0XMI<
zcj&85o6EN>LXtaeS1pzqTiEc@f=nH@-qn1D%l$2#ziZg&glE?liCZK!J0ua5Ib2jL
z0thRda?L$LC}}DSY6-SB>WhZfNF@dvcD`2vZkw5t%*0JOhOt?&l^o&2%Ws<xD=>V&
zBcUTo(-4Nu%oA@(mgzlM6v8x<;bbhkH}cBAxTDsSP4}m^(lZI%>FVj3D|Gs}DwCNh
z?)~k%6M2{h1yXjVqAN2B{NTy$>o}}hJXz?HNysav^sPDg{tm}Q#tAiLAAVdSSAK{r
z)hFc<katq(jA5_ow627g-0P{8>6C7bPh_3Wv!(GZ4g1D>ggYX+W>_qOvsYTsDCvtJ
zpRpaITtIS}_BoozRy9mEDA~kCcZnt1wdzFK-hs>R$iaMR+lat6=~E%A;u_9NkEK6{
z%=W1nCR4u6m}Iu$tXXfjAQejvC5WjXsiu6!D<AgYg0XG$M;n}P(+C=K>RlVaO5PG-
z-6D)+TA73*23f)$UVkd!y#iBxE^_9@TT$GWX=!c97VAgDJ-2~y|B93T+;WN4#}qPF
zTbLp~1;h8g)A*0c^?j{2D`O57k)glX1k1#am27K?K2C4RYYn;x*15u&Qkc1O59XAy
ze$z)E34_Sw9L%z}PK!4sUT)pj*X{l7B4?IV;X@IY9{_DA+K0hFc$wl-k@)ZknI`hi
zt)!wA%?_uzmb24p^~B-$727!!>a}u_5Dm?{vosgOuza0Y)Soeqy(z(FoBLpKJ5#%1
z;z@ce?`{}I-x9p_{_~i`rw^6I+8jR2XpQ*52Ct4m+^vNLGy+%I%H=H->zivZcG{02
zoP2El><x4=Z-2e^l&`PwqUN50hTvnuDh+)|Sw^@;CTncQJf`<H-Hjj$`X7w#$8cDi
z`Q_b0?yU?50N2B3Q%MV{v&AFW{j%kRqZ^{iRNq1b+?tjW)0yQgdp>2vUv5n>SZ;!N
zu&VbjhS@~i9)~Y2x%1OJv@&U)TfE=h1~#}zY=<HuJ%=30OreR-KXOt5;Jzx5B1orb
zOv3^riVgQdN>bE}Tt}>9X9#pByD+^~1%8?sYCj#7uqci%=ocv@A~Uuq#|rj(U_vq%
z9+waQ?8tc8Dd;*qNzMIv3o}{sl03d&zsg5g)hul5c7GXviU)6Oy66$ia_H)(0^YA7
zUAnUfzE2q#_kwLHLF#_A@3%6*))|YqCCJJ$Fxdo$Wq0MI`gw^oqI`EBj@+&vDYoID
zYR<YL$Iif4eXlA09ZEKkTh6;@CZj~;RI>qdSQq*=A#hw;Ze&2)rpSs-E`c-ow{Yy)
zNyF=Am)M+c;k<h4(o4jjskMh;<Q7Cuh0g9hvApdo^}+*<8zEpKY)bt0chN($<}&2B
z3-iPqk?}(rPt;4#h6Tn;Q-)xduauozu+?q5nC}-Abc+<Dn!K<;Mtr^}44yFdfXCvk
zwCi6~dSBys!&EGH-n{2^<ySk7VRgZH)}?H7cfdu+m#YaXTJ-OG6m+6mFV>G(HzSmY
zh&vkk^o;%PoC1tv=Zzg)H>C7k;DRZ9O_JeW>{ddqv%BbhKanXUUYOf^4HTUEP`1R}
z;T1`}<=@S`-b|*4OuKc?t$eEZF<$Pw7`K&KleUq@<VLpCz^tsIl2E5#tG3S+jbm*r
z##655l{unrMi}vh!i>|EEVOH|NBei1j1g~#A)|^s#U`F|$08LrLpHnd^xBp3u~AwF
zW2B_)qz*EB!=Kc(?`eeIz{m-iWB1m(x~vhx=C#u9JUa&jezMGq3<F){5!|FVw!@l-
zMlx3D_Kh0X%C~u9N7UuZQBPVRX1krW<c%m?AIUZe>cF`VL!tH0S#7eewUJ^N{WT+}
z0-MkckixJj-V*t4!=$&b9u0`RQ$xAi5s|U+5nHCFw<>O1mtUR2nv9DmKV9#X6OA<s
z?cyN65Ih`%AFIrB<i-m6Auy<yI*L{0rE?i_D)NSV$0hw{E6p6X2x;#HFz%JMM^E^R
ztJR(N3k<Ke*iY2Au;KN{E3_BmCe!wrHOj&Djurw?a^V4|x1$T;u<av=neya6L&E{c
zNBDGL#&LGZ>E!sh=;XxVogKvfN<HV`5WWN9=WJZzWTxm$e3`m;nwYx3?%(UXH4)qO
zqiMbA<6Vj4Q6;8Wuc4az(f7wbM$ARG5|?e@?Q&mw5bi*>SlOyybuDa{`#4tas0jyb
z*1EGl)*Ee)PLFy-Sain#K#1|2Bn(!}?s!d1SU|O$Gg~3Ie(jKSoUo!vEp~VzkNb?G
zq7AYE|B=@3LTph9hs~^tLp?_3WX|wiVVWyb#||Mw8>F4t#w1aRM*NfBwd(RZrKaZw
zvH9+`Q=B_^OPfpPg690$5XZXGF>ZCL_q=}g^v<eFDd!>+`5+M&?f*61&=9S1Zt<QK
zr+|yQzeCM%QFJxY&>wFI^fsGD2>OLN*52z#Fm2ftp}DuTbS+32SLu;lPaLc7XjdNJ
zPekvX_1}c>!UBRci@3j4YwSLjlebH8{NyXNeSbX!P6ZTId^9-g=DJ5D&w-QFs?V=t
z&oZJlM$j*;qGZncuf7eA^ha2=Ms)~ZkoAf-*cD#*f<_RrT6-O83K3h(W+IpfccS~4
zUb5ZIdx=f`J(8hA4jfmiN~+mX_tE=A^!jGxdF7WzgCph9E65=;_nB7p&kThIh5YzS
z=7rnr5nbkBsSJf-t0-BcOIH>3sb8_72lk2e`%*)R+af3Z1NpTQUd^fW$8RSNSG(GK
zn}xSFF|Ol`^@c3(bVpT}ml6Qr5H~i43A4z_r0$N-t##h#Fm3A86$<BUUn;C8eGt#&
z>11v>GnB!GTt40W@tVX`l-~kQz9?`oz7o#GQpgZ(zf8ep6PS^A`8|V{msPSyWF@B@
zQB2pu;+;PcR9cK=Dh{b0`xCj`m*M5sIF3ktqF0etIQA5cLzt_w4{cGMiHzSBoENhD
zmj5XOA1tKlu9e%qZXr(}y4Ih*{T1iSJSP1(db=}R{5c*&?ed;sEGx5pfSIBW**&B;
z6mToVkp){LEy$kX=eC(*p=s@}iD?i})bM$VL7b`n*z`z0;%ax#0PQgY0u=rb0LOhj
zr>L>#$BcHotuS&{uVp2r_FlxwiRj~EcaX^0`plv8iYB;;@@w*j@;37BI9Yt*z|j7^
z4fXy|%vu)C{l@&_Ziwt)&lw#n#X|{|g(c`QK4J|mwP_#?*NsGqfJ}7>cvRpn>j>N}
zg$y?D4;0I9Zy6>I^L_{se{_fn7C-{DXoKFFYOSVuQqOc%n^qb4GldBinDk=PFdg9<
zIc&opC<Bm057Dsy4Ql+qssA6)&K9G;wBo*B6zF|T{RFtb-)NDQ#>$}Jtpd~AiJEp@
zDGGMYVN8%Z{-pxa9qxUNd#)bMm5d0!HwgVULXP<pW<yh?nipwjPL5g_qCh(V7YjI_
zm;5>f9cB3?Kmp(QPU|m?M$M&oKdiFt$uNET$CtU>X>X%^JRJ+H8iQ%Oc+`kDhfD4B
zokGPNM@u<8nk0M)8>QY`tEFx$quE0T=PTrT!aP=i(4m$8p;W!mt7)%E!*aLr3xBF<
z4gS7Kg;U4Oo2xG$vhnt_%6mNUu6Xh5OwiXVCfrO;$~>IDo6mpJ5ma*S-0?Eb_e7RV
zG&rOmoQef(YP&U0it|u`nI|gV75umWQl5(~eIUYVr1*@P+lY198P4#ca~bxn<5kXm
zA@qC?-CuY(AT8VzZc~82p0B?~+r*ZC_N;c&u;Pz<rElJHhlm$!7lA0{vN>YI?e;p_
z!!(G+(8AZ%xbneW8pXxJvnMRSb$qz<@p1N`-4n;L@_||CtSPMG_pUn@lwB(;J$D3M
zJD(B$aOC-XCq0?^l=anM8qNfWpq|SN*8nHZsTj;R$;mqkzPEA1;u588b(R`6YUg#K
z&zXv1>GiMT_L};})h>1kK%7R^E<jbDVdQ1Kw+No1`vdPvz3_`JQ;||!<s?ANuCI5+
zi|6FJdI>6rU$Aa_WvXl#2LFV*Mz`SK&(@}EaZWQiof`k77`&NpC-k~#hTEk2N>;q`
z`0<g#Sy%7+cW9NY*8};;xgrW^Ju@@-&DDuSXi3bq0)o69%4ZjwJXmmhnd*^k6^XxY
zTLO(R<{FO}N_C3S@_Kk&v}>mvwibHq9OkN@>GZ#1$0|gV{Yg8XMBG&C+_&%G<FAHH
zp$njPyaFwou7PD)@cWha2QoEN+|E1>XYJ!PR%Of=vm~_%7Jj{~`W+<7-({yF#Mh+R
z#G2z+VlDhUS0*Jp-jhSOO((06t2dRk@A1WrrA#5$rx*KK+HTmKQ}IoQ-VlmWrL|d*
z@kU9I^8cF-LL7!<X_p=~mMp=-+T!wQCWz}qLC{8rdw00Cmz-8!w<bqR-VLzED@&#i
zznobB<wEYnZkUIuyy9S}YyC!+=uE3J;=t=eD6(#k$zrkaDRNt4qBk0M8*`Kisa&mI
z>Saf7x3~Z<>%fFFcm)zv7FyvXB+Y5dWRPOtK$o#o^n_BVplUW@Z`){&A$41o@I}Y(
z1vgb4sA#|%&f(Rq#G-Z|rghI3)U37%^NG|f7rOuim3{C)_t!-CU$k2g7Y*&!XFFtq
zDz`D=>#~4$zAAq2>*r`;;~T0W$QCx6=?1<*&o1xDm)s1|#04GygrP?%2M(nTefo2%
zpSxL&RpaRLMZs1vGI_BkORuOpOb7W;+9`P!%^rOGZKnvw0oVbst*_4|yPJcDg^Q!q
zoZRVjAB<7zI(QC*bXIR(V-=qtf!yzfJ)5aSYoE0Pg;^~2iN21)rC_?dkQ(j4F^7Q@
zbM{}g8zHt<Cp=M^F&y2;py(XnclJ}538L;&v|5i|n(ko+Gd4Ll2S10>!EFF*o;8M3
zDr01*)a)ZD;RV%;@5U!iXR~hT578N4hD5RMuBLB^ImvFuT;UlxPr;T{+#Ttf{f4o(
z*fr*I5LXVXym%$E5<VFINP0>PF!9(vz|=N39>E7e#WGlQe9hkyOYU{N6ZR~dKhhh)
z`jYp^aKv2r`Prq`TvjP3QxU$mA&=z2*Xw^g5uHYglJFM$m(a9O-3gwOMp&L_QzaAq
zohC+MZp*M8TcU1+wO)jyUWgOl#BV2ooJc7o>zx8@Qgg*fnODEQ#Oo0v7@+DNWf^8I
z*1QK~oJ14Nqz9jlE_@H>E>L*q$PwXTdYJ3G4Se@DS1Rx0C@=fL_g~B|XBeo->wGOV
zsVHzmBxaa={&2VP;nBZcNC5|4|1do`9weH20VKJ*w!&&E{x}q}uFUxFZ~3RTL$I%E
zK#i!BB$lEkpN(3&`n9#VQWRR)do!XH6ztvX2akm8+^29h@q5#G5vo)Xws!P}u=CY!
zL9C6KF4A(>C<Vp`6Ux7~%_Qce#xrYmdz@v7G4LU&4^Gq<NtZfEYVdE2T5;Le<6mgb
zx0O4NJ(L0I%2e6?FQQk*!Iu6n@zOmbjv?G~68%2#n0e3D&z_&PVynG%VAgwc#vT4j
zq;fNHp?ZwuUX57l^9<WDV&3{_f3T=7G^m{``j+aSbcB;2TD!6KQT_?|@TiD~@WFyV
z@?oAtpWcKl^!JtZn6FnaT6S&adO6f%5X7eit;q`k3fuDxRq{~nP|WA=voK@Iy!7cg
zFSKG<8%QfsU2pb(R9o3s$hwa&MQbIddX>9TMNuDk11DrXE@Z_{`9C;(5MqvesF$DV
zSU~S@-&nadXPIoo5_RAqY=F6SF8If^54rcKO%Gd*(Qt;le!l|rLKuwtj=yKy4~cyh
z>^1-SjIw4;(YmCeQ1I!Zfo&1cUL`<~DD~M{1Hiz0bNQ;&_yZM`Xf`C*LXN>z$G&$^
z#+^?VM0>l@oH^FYC4Nrl30NbK`0@?Yn}K+I-#OR>c#lSFJ)MD@?N$nuT4IDvZq-z1
z<v~y+ZHxC+<<_YGm7%ZOV=koOm+Oq{Mc40qw9jfySF>6Q`vjbMkcQg}^vqs>dDj07
zE@BEmDo|Qd-HH~R6hCh!w}Mz>T(1uLVZ5kkOK&E@1&nqs3t6#)F}0RUgZWiS5_UI4
z4vc!=J_b5d(i@l2fQ#?;$Db>$1jTGGaE3XDbk8?DBCq0*bTAU3Dk2q_)vWm(TqulC
z*qaWLbQK8WE`anHh~aPo9=Wor&a(TQ;4o?XCtZF#8}-pIsKkUxATdqy7w(1Rc{O=3
zG_O>C(g->U*&-{bfI+cq=J9V-#^u_Ge3J^1SU&ZL&%nkC>lU|aiOB6J)6R`y22qB^
z%tcQ`?!<&t_0L-jK+Qr7v)J~mAyAvg+c00nyWR_H*)8;D*?kH+y70cF^BTf?Y^)cG
z54Q@<lOfYmV@qb7g14LBGPKFFZkQBlBbbDR|FqFYSQdDQFJ)EO7OAXr#b2ePbYs99
z-`f@mPIkEElCEsx?p6*Y5>qfWGje%OKbPNMve=i6<dl;X9s6)sMMNbckUpC8(*2vL
ziGzgzU1c|K0pp1&O$lwDmeP<NBb&W#oXpKiP>T1Bmf5en03XpCpkmMAmH)V{TxM&f
za;baM{C$vZv$Y|_4)_ljfD&CeF_vwx1nv!tsK1aGyusg6nnq0|{0Y$fL^lzr_oBYN
zy^v`(t>S;uU;A$&`v0*@Jr22iQRg8)g*)2UeYDtG6Q$s@{ln$^6PD~;{hZk`NNTqO
z=&#JA(5UL6wWCOug7@YdCc#^8G`KuZqq#t?O8`EglL}rgC{BQ9r2w9dZHEYf6V=54
z5}QV`iJy(_=KKp>_UgVJaAa+k{{H2KJluVCTplQ+62cZJ`Fx9dMN!e~aTn5klP%v>
zc;GOl0KG_nbSA$tQ6mNXA@4yd`=bWAk>0FAtMCc})deWRqJ-z#w0!g4Gai**V12t~
zlBZSLaBlE?wsPrUNDOOVwpP2agk^%Fuh(dux4UtTn`7V4g=LuA&1)>z?~XgS4Dc%O
zd?CwB&Iccd+*m7vd5#m+mtrAjmvxA))sJr<S0=9gQ8jqU)7g{;g^2J_v50&X78a@e
z_IOmm5qQ$#4gvs(o2G*mQPK~-DTxFhXOC(B1fomdK{yfrb>?W5aIF9SF0_ZM22`k<
zMp#9bf$HVH5>h0j9>Iznu4MUhm13drrU!E4EDq%LoryJlp%PY!X@wlZzz*p<Y@nD?
zrWV8kxU|yFki2LKw@aaUQ9-{^F-w$zU2jMmx&?vZjd6QF{Ie)iuZ66gYdmUO(Y8o>
zD+>wp7>R<5Cjvd5&SQF#|0um~2m{}%O6j#YaT{&NtBLaj`gApoKo{mklPiq*;s^1i
z`t{co1v#xGX)(CJ_|nF$o?gqgSLdsO3My}1`cpvp8C`l}@mV)D!&x6V0}NV4b)}AV
zh*t!8GQ*@57;e^l>VW1n+ZqgP-?NY-wTfD7QmwR!^&qx<{aGluE|&rOQ4LcpyYGB@
za4%$A*Z@FTY-O@eZUXpZJeNZuQ8V~xD~I#-YJi8=&S+YZa&uCy0q*@uJy6fqD_Q8B
znl{!`_vIn$$intibXR&SaCg0){P9$3`jMOp_!)zM9E@*=;Am|3Yf$`(#PTP*+z?5Z
zTcoyrEBY#{Q}-!B`8Ig7VeNaK$yoK-kHHGmB?UmaEkN=$4P%D;)x*sCIocgdO+<Pj
zdD8Y&Rr{6q>-_#}{GLhSP&MTGDc&X}J_QJaE?X1sa}O}D^tG>Y6X!K(cb|2<?*q81
z?MI^)bdG;K;R1d@CF<V42!wSgFG}#Fl`IYr>wjBo!T&9x?=Sl<a$RhT?E11xfddo*
zJo3OqE}CyvQx3un6!zGiDnUpx^55NDDf1pH{h99l1;C@f^Wm?f7;QL<sCJ?hLK_qd
zZGDq<-hIW^?f19J?FN3X<eLJe9?%koS|32$L-JXjl%$+E?u)-)@melwU6buO0r*Q!
zyU|iRJJ5sS`A6LH4v)B%e^C8eWGRQ+u6%en(>9uK_($aO^!ppiQ3+xP7Q0R!Roh=@
z9GH3}oq0yp<!C7KY(`e|Z!D(IV!8jupWL1P{M_z<LHU($4gW)MY6R5$FxNqJiZ%P=
zH(Z+k##6S#b-V1ofpMLeYxVRC&9!ZYo;Az%7~#T2^^5+bJyQ{*4{6q5M^(EnJ(K~S
zB2y?wD}bUiZq#&XqcThFJ7}rUVmN!<mYYMo8sU=j$LRqG>S^OKKJyh&8|81bOxGm9
z?`6^}t*ykhu+y?fTW&6o*g<ywqI~0=im(`F{p{;s&bU|udc0~meB&AE>fcC@zhi_f
zW8<cPMCh>qESoEA$1r6~4f;R2q5+<XWtVjLY#l--H+3WTfaWmy1rA>y6YhlX@A-K4
zL595;CVF1TQneC@d<t!3-2+-<sR@EL)4EpRVF{c4WR;@oHUS!};Jx6nWP(me@xX*`
zWO5VFL1E{-SFEWIt~vO(qc$m?_q=A~ZF(o%(aC<K_0$}FZJ9H^;4ihHnBy2ZnBT{)
zqqy_Ch>iljA22(RJ<hcJ9jRyk?|AL+;dRK6RwjY%pHm_mghd|k0ARich0A|oqtXA>
z(co7XR_}q?HL(~31}I49K>pM*KwRknH=hn*NjobV_jcC%Ky7lr!n_FDiKf?}8VSlZ
z(ei6oPzp}tl@b8dOL%R}`^ZU@Xd~W3fd5Y`l>;z_|0J&r^ztePozaYW1JYvl?UH*l
z5GX!?`%n}r8wD_lrz2Tz>qG7&ra8P1n8e3Y`BR)nimgG2k|;SSY~Wz+N;(WbgD{dv
z{oa~{>tesW@7@%5*OtOYw@Q@|RImH$!I?73)JvP<Bpr=+*H!H!fk8K?%zj9+?q_ub
z4w3?BJc-{wKWnVEp=Os9iadEqU}e0jr2O>EhwsxbbqXfD%0MkM`>^-T)hcTG`_x?W
zm31ap|G+r_VO86XgT|{e6UU79D?yaNto&Zst^_&N;sTstZy@JzG0L^+5jn*}wnrLu
z9#LjBZZkW7;1F{Qh2J|8E>%Gu8|T|0F)Tf0d@aSSVwS4LfP*%!_w(KLov^$2vj!9J
z#Xs``{$HDF{GiDJOd=1X6BgzD9730gV!v8i9`XaTm{+I`fi^?#-1>woA^#L6d2#;6
z%v0^GD)pkJ{Y8s`x+Ps#_wegCe7^`t3x)f3J-6@lye2T8$h(z)4K&ATOHF!A-L<;)
z2P*3LNzCTi+KK%nW=-=FARhJ~N!c5K{&jz(=CTwATE+m44;9IU@+TmBa(iXaeV&#Y
zfMJnmUUdHgLIwh*=MIG*WjOohrMB$(U`o8cG`4oVcXgu1*+Wk#dI!Mkto1H9yinY4
zZ<|BDX}#G%&9WMZ`n^BCH$uumqWLfC+m$v{gbV4}*U1=oPrc#Cw?NwRLiX>CP_F=o
zDx+mX=RZRJmb+GWp`|>}rOcM_Sr4ig|FL-Va>YN(OzwVsaqoftUw-a`rvpJK0QFzq
z_4t6Xa%VR2U>QI${D8{^1XmWCv{?bR$=*f|T+7hh(>ivkKYT92l^gH{fat)<8=AQq
z0GcB%+8U;j05xL@h<ti7=2RoB6@T#z)C(X}yA9;dB#6i`;hTAY3Afm@BoMLwsY5%%
zt%@;2w}CC#F|yGSjZ%QyWMVusBIZKRHx+wOH#~=abU+x8Jj|SM7i>C{4p`cDks6`8
zB{S?b5NIqMBVYM#H*#kKu2{_XJC()fB`8}g_+q^+s)6^r7r+U7`#MV;rg<R};bR5X
zOS#j8*YJ81L?X{wZ%uk59D$!p1!9+&Sn|2F;lz#6d`Sj=?XT87j#eN&FZAo{XhAmb
zipJYE$*<c%(;e{}1^h}hoa6zqG>%t?!bHwO%F3CB111V|NHH_@Jj|Gcx)IwUH&RtZ
z4Bb8sPHL66IJY_mIbKx`)3wKncI98tmd7=X_8J1qJ0%<in8LXSd|&|=iCNwrr+9E0
z0$|_HF?>{sFr+tAKL!rUhTfgrddOy`CCe~%SCw+Dtk7533^<t~uG+38f{oLSpKM+e
zq8g>cRPWFmu7|+BL+pUCqc-3}m<&zzOxEuAOq_1{5ONugsKXDAxA7ezMZoxs&0!n|
zZMV|b7$2#H+(}2xGy)&St;8i$Y82a)smnkIpK_$fq;064yg&6GTfVGp^(dv11w?r_
z=%wW~@IpN6&chsh*kxIb2-fFNFs<=yf~AOQt6hiEG71h{(XuDm`I=58H4|VOVF!z)
zpr(O00yH87AM5$;%?A+XOhJ+;_r1`W=u~|^TA#kJ0@Nb4Lg1^sdm%&i8}pgbvmd3&
zev)-Jjp&V3*#hsj%xH$koDgH;0rt5G^h^U!0fD|9QBT}-fh6g?AZk!GTlk)C$$6U!
zDF5oc;d4`Nl|Xb_vo`~Mge#!p8xlP-by)hK;aK-B{S8&BFhZVZzYerM%tehYd=@k<
z+RWrEJxfzOhH55XLoJ5WyNOIwW~w3dWI!CIxAfU8N!7vvMc}?Y83)^e(#rkdpwee$
zW2-Lt*LA+F9aw+=XEDZn4&Xy;`n2G~&kE|+Eg-6p`kO8A1OZ*Z_R9oq7F7Z9UiE7E
ze)8%V!6n5~F{arUra=yIr5ob(*Mzp`xN|NClc+@Nd@ktlfDa#2eX~7v@b%Nx{;KDY
z<4YE)G+cy2Jt6OjqSk3w7bdcI+R#2Ceh!RQeBMs(zpWx7x7HeK-f3MVqGa%zMp)8u
z^cAy+Y23Zer1%`bS7hOrh5#`N8mv-W8jnkr7BVb>X4pDVN*qA<S`6B6{W^iRP_Eo9
z&$+dZG+erqYhzyldzX9DmY(MmXh?%L_(6Azgv)$4)cn%;DF|TC98i^1fTM~2<UtOO
z_R9Uoook}4`!Ip`H*JgVHa;!&-zT~3%*5tEEY1S(2YR5e>A|B%)<R=g<=mF903yWR
z!iKQ1U<Y&+sD<YMu1b$vq^!+uVEuh^!9^8mzo*i*1_j1ks=wlKwu`4DxfH6)PiKL{
zXX$0CXC=$I&yM|FvxRDxBeYyeDk0qv2N=+p>`X|VS;L#T(93KNmp2AjngzUriY1mv
zu>Qdbv|QLi>7a^ZXY(NC-&L=x^#6PD5xm>3|1@WZ+slxk!jdm+?zJ#e`^ObUfu^6c
zv4HPz6A5Q#`B{oI<eBYEmN=i&lirNG>?zRXTSTKY@RQ*x1&XBjgx6C3!V682FIGB1
zm7iKc%<gS_Ex5k&37|LUDxl)V<oB6?!l``*(uRQ~OZ-xpI+Ex|m@cs!E7_MqbFF8?
zwXpJdtp!H}R|0~(v^P%7>b`S+-&71UfC%aHXM0~CJPj9-PoeSqT^4WbbBJ0gmgaZJ
zIJCU_doxNJc%i=geDgZl36H7B;7AQI02CpPSOO#f==5ru;eX=JHr2eY?7#UN&G3@~
zP`4P2?G+ZTJ|^^OxX8+xEmlMLJs`|11TF#U<_{(tV70IYONbB#$e@i!naJ1@&7G?9
z#0WHQmU;de0hm<(LU)T?Cj#%Z%xKB!Qzx*+UN!@n%M0{Pt*wB#l?7y-H;^1PmTs`L
z@pP5vReC7LUlKcppz;_!DtxYyR_ZT5Z#6vdRQwF%|HD%dz#9E;p1bgGp3#fLSP1Z;
z9Z51JZc7g!#bE1_HXc;GFP$pCKi%Ag!c92>Z4NY)Ul>~-u<nn8pvG#5wkUsN1xo5G
zz`>#5@c+o8_NF)IDB^QBhOMHrK<hZDwjomoRNyE&2sn(Eo?jU*;sawMgtd}oFCElA
zVA_Kdt;!!f=(S+x9?V@JfkR|kArM9>*|imSrU5N62Fy<)kQb_*F1-aPF6&l+U4zO`
zq9M3kV|fXysg}qEs6Ew6PxUrh7V~%E1yl&vYur}a*8hgJxk5_h<lI-kf`la$PS2+f
zxVoW23$)+gZ5%)SG1r3Vdb})W+FqBPakNb9yktN4rld<z?jVFo%nUNqo16OL2X+Kb
znUp(t4p>dS<GTy=cCB9_z`9lxTwmH8Kppyt_Q^37r6H>vVykMtcjzIx_)iXwpZM8$
zvNuB<FMRq|zVYk7t1j;$)g?h1X_)}1Q~f{lq7Gx_!m+;3xq9J$Rld&v#`S;=g$hd0
zSq_z4@^H81VN<8!4?M9y%Ygsj!2mZOkoaxiF9*>9Z8fY&=o6Uj5$lGXOl}OvpyJRK
z2n$)z*S9V^)f(OPu@Q7e#+%q5kQI;yQ(=+(=9gn;Y1p1Nw?oPraM3bK=S1WPKN@N0
z7E1fHHKDe}^^K=ewLA)d$n>nM#Ojcyb=IU==v5gz<=4M%043^84HhLw+?#{z<Rm@3
zMwr)tCuqwqycs>&@o&)=tyabx6#s(S6z1+b(0G{sdxevLq3#YTbP`lSfK?s7D_T9D
zQQO(7xLLFT`dRx|?bgT8BuQTYr|w(Kfkbn)V*H`z%48IT!^}`;VK<)uRQT-@-QC6v
zweViYiiukQuhJU<Zdlb6)Dj0E9-2VWx4ANAZ~vwgq@)eVBT8|ys1t(cC_r52j3Mgl
z%3Ho$?tn@kpoO$i=pGb4qU7b3_W=0W5F8R5%g~7)3AD*9&?+bek(i?UvHm-;ri#F4
zy(7c>0rGu}4atCaeb5Mzp<I^BR}Y`0_k%U`C;^o7<K#B*oF<Y7F=$Jtmrp-{0}n_a
zVASJ&8a^HlKL8$)(u!Bx6Ij34b_m1*3c)@cv<KD#PYc9*K;DXki`Lg1kHT&y9{lPb
zDO<HUl0!wH9g`;(IJ_-IsCvRMz#;N=9qZ95cVK}%aHg?9(n({e7yV<^_DEXad6lv=
zFkN4O`U8#$JPs`ed)Z?dEN(-S-g}_?xGL|e3lCX9CUk4lQbPtA#dK|g7{prCc#fbW
zR(T<-5V!^oPUtQ%F-|D8g*5yu(w6E;OQy_WAEm6m>5xA-^q>@y!Efq{#i)sutMOF(
zNgO$u49fl^Xlh*CyzGfS&=9;<92ebsNXY=ITm<Hmu|77a;?JNEGBxQPyw6*}al(-9
zybF_25Ty)8;4~OUqqWNz=>9Y=x<~|;_U_kXVn1o3(8Tyr^*dC>(XcvI&BXYi!E#tz
ztKZ+P@0x+M433+s715Oub~p}vR~g`y_c-u<5p3dXQ&x(r6anv`s-*Tv#uF8*yx=%c
z@g*RF!Gw+JfbY;qmMt|(<bZ);nFYoL)OS@(YNgrdxGdlUa26Ox<S&!O0sDl!cj$ir
zxc|!d8qfWAJ62u#XeHm+WA}#Sn|*}H14viE(tGgA=;4;i!ejgT_=1`h|D4@&#pE1N
ze<QhJ8&V21!ZD&2DF9*L28Qupj64sTyYJHzs)J4Qi&!VW9CU(;w9{(WoxcA`Q<VFw
zeuy<P^&6NDlMM`kmp?(fc(B0C1niZ6#m^JXT=qkBF_lC0ik$Kr7MVty)mG78mgWV(
zw2&LJrQ7{Eim$o6my29FW#;;U3jknhNGko$Y<+hR1c;A3rr`GFjims#X)p>P8W=3v
zEzQ9EgMUbzGUK4Ia+a~{ePjCc%S-+oX!a2uFC;buJo@%eYKCv~hO(t0%(P}Qe<3qM
z^7=O>3fk<AhkLXzLMRt)(?~qP?t`<kLqD(UAKfA_LO7USmD>?}0#uJhXCO1FN$=T$
zACPm+P&gIXOhs6PjJk_mz3eBdU8^_%1t&rVdpAGSwXC(jG6ru8bY!+Nl#rc`ySy+H
zOPjsFWrfv7c&~Sl-hfFz{BurapEytvzyqUm?%Z$JCJ(23r1qZg11t@0=cyTPkasI<
zAkQ!dw8Y~+;LhFaN=XJGC~o#^d>=UZ-{~}97K&*{U7=k_Js{O=TVB*~x5ZvM4%tv*
zMeD}P1*IToiy>-jS)Mj>ujLPW14!GJ&=7`k$ivVGIcgyx=B-}l+6n~ZFFXUKAhvA(
zSwTFmVqvd{e-5~T;-M)+^S{z0c;q`TGNTnXjZk5&iJ+BY128HxnrRTtZ@2y9`vT~?
z^7PvN9=u*`v<1rZYY2}<_FmvWU3V_MpFHWc_J!Zs6U<1NJyE>|`oLz^Ah>?UMdLuU
zxL3EJgMQX|()XdP^YBr-w&9n3-lWv}w)QkZKP$-m0QnuVRKl|~V`rUsnyoeXZif6l
zRTqnQ$%`uh3+cjY4Yu=R<N2JV7m9?QXs%5%V;}`n{tJx$|IhRb@Yw(8(6oOOy#K-`
z_^Y;JKmmi~L_@8AjonkrO5=dfHvz!gyZUv%D*c=T&L2BjXZIf|ef(`(G2qmI><*3D
z|I!@whx1zTs4%aC^v?84`wDZ%KM8~<A%QUZ_#X({cd*y;|2-b}Wj6k^SK&Dnp9lW}
zkE2f@o1_0F2)_z^QST?l&KAg?U<SdZU$(ByX`&i5h(J)!%js*JSXtE|6cm7!+K}*;
zR_IfAx(c*k!K5|Zy!IazMZ1ZyOE&6B(;p6rNI=C8$1wpKMjH^H*TG}30~=4k*!Vi2
zaDlE5xT~3B(Owk-|6f>CfOYsklJWo93=I@oF#Kf4?QiE-35U9kAS*&J3ur!MTn9Rh
zCSXwf%nuw;9*_fS{h!eA#$3kNQ{-i$?dXMhW59fzoVa>0!o!(PZ+n^_9^WZF72#v>
zJkIiKc)MtQ?Vmg_e3IE}O#edf``@)d|BEq&8r{JD{0v~==mVv}3~=iJ026KyRuKLs
zHCqEtsGZqbCwUs!oiu>38t6}agys>woZN1a$_6C1h%bgA^$L9Fi%u^v!`-*%a00h$
zp?u6KZ{iS))dOsz2SGl&W9Wk+F$dtdYAz*lay?~h`kN)-g%>~<n1N%yp%CaWqu>Ce
zX<m09N@MW@WBRi53aSO9(9!1;i2H!kehZdX0!f1<VE!20pFG;_O7!_>Q+(FGxJ-J^
z4aAg&e#4-4S2g`{EdF{R=PmX?lg&U2fPu*&*o7a!$mw;CMj3d4MQ3+x7rFJ5nD$49
zP;SH9<UJekYM5#o{Zf+a;O(7=-IWToxk=!oJAhuRa>$-hJ!}R@aUIBD+Z)QCVg%+J
zLWfHuM>f#eG+V)HqnbY6CywCbqQzIKg&=Cy3)jptmPlx2Tu16wgT!0g37jae6G<9(
zkl>R!S7;oRHV}hd_K90_uB@zKvh+NvL~QG8L-U<?=x8hem{m!j?gW|nh(P*p5+UN#
zNIH6oOwLdxKnLQuop(mM{3KIlqIaaHxXzGkaOkR+kklcw^H^8ZK4=+lp8@5}rYSS<
z6G;$lzJ<m}H!CIK*iYck_`!L<9RuD(sBr{!9Hhyk!fD#2??avw;T!3+12MIuVN8M#
z*=+hVMZJ)23d&JXtJsd>krZitu1AJRm!rvh>vwDF+uPB&DEH|*RvYs@J8pNOpaTlc
zya~%we^{soCBZ2zwm-3~VOvH0ciMgY0biefYn&E#p&L3U$2l<j2K~8XTaJMu(VsZv
z?`u@Q5f@F~2XvJ8<w1|mws40u2v$D^!DH$Jq37JE$fzT~zws9|Mnfajs)QIH4T6@@
zUNE8HAtzL)7x1f@9o1`b1W-%Wmdq70tfLkqJTm7GU33`!Fe+?Y0^m4@Y4hvL89V+q
zfez~75w`IRR%|XzQ9b)jSk0i@Ag<;CL&3!5Pmlu*vU%-kNlm;``4DH>JxFIgegwvP
zC5Tz~Q<vXniD}EJu*ldC4z*v;kn5M0hsxr*_V>EV&J>0;@V1DA#K<-qXs{ZAt|xk@
zW~@eOLI^q%x(PC0S-~77nqmxzaJ~ga`M1ke_)$+9+HafTSHy+-oTBGZPcmaMr4|yq
zvjpl3z$DNMBTrITGa1oYQiB!?;Kfe)oPRd#vdsQMD(O5lm82ZhNWZ23_KGf0;Hm9x
zdWsZ0Inu3vIgup%DU@~UjaZt;MvsQ`GXpv!T~R-&j698=7rcJ(>ctr5da<IEM>1pq
zj{t-}lMJ+y!jo4-Rc(7ltZyu&W_atG7CZ#xO4VyEiLK~+`(Vh~cWB$v2xRzyD4gZ{
zpxN?ayeY5`Cv_jX7FIIDS@!mV(p&=rxuE(L`pdT@)}v=hJ<ITf{nn9<4`3^B+-rW#
z9rlQ6DF;%U@gOl9vO}KRy#Rd<kCqV~N_u%liswirHgW-VVu&Ua>icoT=XR)fkc8vj
z7fW^<wOgpe`XW#WNKZ~|*x%;f=i~mYJY4Q*BGMYc_EK{`?5DHqZ<H&GFj#JfnVZ>u
z&=Egd`GLxF$1Mak7ilxZw;Vzy^BcHzu>8!dW*%eXZfZO6zv4dCz&O)?AH&o>{m)Vd
z3<i{p+IunMMBVCdqloM!r(N^{DqjbJr$Adf*c=eiklyszpM*T=GGG6rx1D005eRSm
zm7oxj1(T@4?N)W%GZg<q>*(i%-8scz3@d#$OOlxSsFC`4<`I04Czf+(N>g;M*oXyu
ze8N~sD(O<r?S$RF_n?#$Ux3%H4v*${a&8f<`et)d@(-f70QXP+5_kgvhbeyGrn0C<
zxzTPY*mT&%%#t>q1O~tBZO6f30c7uhtNJjr{ZN4x)^Dq(59-l^5_|;;pI-`e?9r3m
z?%<5n9mMlfqB@?=ge^GgCYZWq#6S2OjP<Pmr?LEx2J!<BG-M?Lw52u}iqN{g293vz
z?KTA86@tE-)VD7bTZql*<nW%CI?K3p2u7{?`uKr*1F?EgjvamR3Pgk$Xk0B+-i@A^
zXYWktGnClk-<vt#*tbVpCrC^WA7mEgBt(X~f*LO1(YxaGUo9-!(mb(Gp7UvBPt+uj
zWFutkwZTo&AAF5fkPANX#44KYRgN-feE^R8-@$29!2H8ej+?P1vy2j3IsAWVf&X8f
zoqIHtc^Jn<h0>J78BA=tj1qRDavd@=x~!xfMTWE-#kNCA7g<JZVn#{DC|QGCmNd!8
zNMqzk*&<UYTKhVq#8m8{av5fy-?Y2u>^b|#?*2LFyywh0?>o=$eV*t0`F>}X+WF52
zibY6A>8>!8rl!`4f0FN|(^`XQlM6p05t)g@eX`EZ^~5FyFIF)Eruf57>!+s;iO<7H
z=X;_{epRq~+Ow61OX@u6flz4&04z;MlUQwsLk2S}a-=y{Om_h|p_@P$daO%1ya)5H
zm5{2Y*hiN&Bx|jlx05C-J72F3rn*vVb&gjTFzAckK%mG6#N{HOr2DB#{zAxtfu)}}
zLp`=Y_$th2%)WSeB<(Vux4y|;o0LS0(kAiD)YLA~yV}eU9uesL3MG@(iI7ePn~TB}
zJz^2sXN0tIWr>{p4zMCrPd96=h1!|i4bx7Lf~`tStbK@9SFjXKCo3nOs4|5+6Y5@H
zw+GGkU;+G?l6$46G)=ZleHz~J2d(4s-9ArLf~?Z-k(Q^X<B=g!L6TqI{6`${Kg2Mr
zn14ll1vnj}!17e|r)0Jr+PMIrY^!4Of{`kI^nhw5-BwXTgw%x{2txilA7m_h-&IL0
z_pLBv{(9H`JtlTh)~XaDxXn-lHBzhyvJv~I4MV-#)7@n{IbC(JF?Wz_0R*j)_b<|-
zp~uKZYEJ|uuM*=n0Oi5B<zmz8#sNXH7-q#H#EVq6I#@V=d=VT?dC-)4cHUWOy!7Ks
zuw|0R_tjYncoz2~JQNx%=!psu`05O$HL-6DLhDM)=kO7nVg;c^4N(o+M1iuly_7XD
zD3;bsuN=J5Jc@H2^dNXEMz~@R0HpYqVWYRJMV8<lynbwccVpb}E}wu5jWH=!=r|-?
zp8HM&3(pQioSqWrjIwhW{1edKj7B_xEQ6U$TO9sS4+W(gDmIUU56wZ@XX5w++I%0N
z<t<lbIBjaQe{+@2P|tZOj0~PRTQeyu8;y93LhoEA<!Q0Ajl8U7RKdfP#1yU7Y!=td
z7XybZ({kX2W%Vx?SP?u&${AdX*!si8#2pTciLwS1F7=Vc4I+Ei{Q^hF1lgQwA#K7k
zv*5{QAbC>-cwZ5r8~Q@q30#TaR(D^8GL9}LU7&?7(Okqau-t}{(wN#v2h?(Vt6wfe
zb?)!BZ04J^PR7noFu(@RJTP(4E0096j)^-5w!ND=N3g$jZ2ldC0R2ASpUCa(iSg=f
z$4_3-FHd~}0UepA3pA_*$(L@*dq;t^YJJUTO~3VbbM8qp4@R%nl|G_;exRJ8u{32>
zrewtI3cYEQ`<Q9cxyG%T_-u<~S_Oj}qKN7@2G|>51hNI$QQ-*JUbL31(>%FmCYBvM
z#Id?la{Jn+xga0|>efxDPuJ_qqv{PcRStTAc5L!9e0r#_SuoeBzy?{%3Q0RcchAm~
z*Demur5cKD66PD}_xK__0;yAkL$tP|Ky?Xn<4+Ay15(0ol)p>pzfi0eYd#}Mv+-B{
zwovKCP+!W?XbX)q*3_3=Unj%avJLDjC?%ISw6Me5B7ZXsNHKG=ZfZ%v3<n+_7f<#0
zBOazzxr&yoyNuhLFto&HAa#}<Bh2-*L95@a&0iCu+_5$=X+9lZxSf;8wiBK}3g)ae
z?VNiX`(#wVaLIKoq0!r~Y?W#KFTMCBMD(gT;`2P{i&nh9lV2(spv;N50gu@Xd4zYF
z^{7iPkfQbbN%ea~L5J=N$MnENx;E7PYRwv}Xun~!s|nf6)eY`t=iCc&m4`MhrB^J}
z3y}zndBsxSL6EWF_q-@dk~6;1+;30gN+p9?f7*al$|7uOkAYt58BkaOliH!$r{=V<
zZG0j>ls1vL(shqEWqFE(Kllp)T^k)Il8|0(%~=;t5y3v}HA&luf4=(o@LuktLW(Z?
zSbM%(PM#w=Ovs%>oXq$Kdt<gR<6Q0fvw}x>C&Uvi5%4|~U5W$T@69S9utf4QE4tP)
zhzEPTn?P;E_5HQBDr>t6Ypbd%-uBrB^6U{<*b|?pmss2%Lz}7iCG#ZtQWRL%4TUxd
zBSaS5JM1-!o6n!X6fM|TbJx6cdm-Y1l|cp%rpbo;;;oVKy3?)7DVB1$YJFY{ggW#K
zV5T2~1K)Omk*)ceYjR_~sRqdcCM@C56(~By9uU~r=)hGo0c?lLp!@8*rlWx~k60C^
zlr#AmG{M|dEThh@$mDasr)w<0uegSnVDY=7_JdW$UJ$y(F*!DU#o;K9jbBcRsa&`M
zp1uvQf`TtUl0V2)pu?2vlf+wHQGmBxtH`DvkiL3^sM}8=Y%I&}=Ao^}9q8?sRgkBy
zw?{6yCqgF&8NSIiG0(Pdgb}ebUPomlmK|xP|Br_5U;YJvd7;uYiR}8cxdU^8Z&BMy
L^Q7Kz+n4Y==>Y-W

literal 0
HcmV?d00001

diff --git a/docs/configuration/env_vars.md b/docs/configuration/env_vars.md
index 4f37c185..8451c7a6 100644
--- a/docs/configuration/env_vars.md
+++ b/docs/configuration/env_vars.md
@@ -14,40 +14,9 @@
 
 **Performance Tuning Knobs:**
 
-- `VLLM_SKIP_WARMUP`: if `true`, warmup is skipped. The default is `false`.
 - `VLLM_GRAPH_RESERVED_MEM`: percentage of memory dedicated to HPUGraph capture. The default is `0.1`.
-- `VLLM_GRAPH_PROMPT_RATIO`: percentage of reserved graph memory dedicated to prompt graphs. The default is `0.3`.
-- `VLLM_GRAPH_PROMPT_STRATEGY`: strategy determining order of prompt graph capture, `min_tokens` or `max_bs`. The default is `min_tokens`.
-- `VLLM_GRAPH_DECODE_STRATEGY`: strategy determining order of decode graph capture, `min_tokens` or `max_bs`. The default is `max_bs`.
 - `VLLM_EXPONENTIAL_BUCKETING`: if `true`, enables exponential bucket spacing instead of linear. The default is `true`.
-- `VLLM_{phase}_{dim}_BUCKET_{param}`: collection of 12 environment variables configuring ranges of bucketing mechanism (linear bucketing only).
-  - `{phase}` is either `PROMPT` or `DECODE`
-  - `{dim}` is either `BS`, `SEQ` or `BLOCK`
-  - `{param}` is either `MIN`, `STEP` or `MAX`
-  - Default values:
-    - Prompt:
-      - batch size min (`VLLM_PROMPT_BS_BUCKET_MIN`): `1`
-      - batch size step (`VLLM_PROMPT_BS_BUCKET_STEP`): `min(max_num_seqs, 32)`
-      - batch size max (`VLLM_PROMPT_BS_BUCKET_MAX`): `min(max_num_seqs, 64)`
-      - sequence length min (`VLLM_PROMPT_SEQ_BUCKET_MIN`): `block_size`
-      - sequence length step (`VLLM_PROMPT_SEQ_BUCKET_STEP`): `block_size`
-      - sequence length max (`VLLM_PROMPT_SEQ_BUCKET_MAX`): `1024`
-      - sequence ctx min (`VLLM_PROMPT_CTX_BUCKET_MIN`): `0`
-      - sequence ctx step (`VLLM_PROMPT_CTX_BUCKET_STEP`): `1`
-      - sequence ctx max (`VLLM_PROMPT_CTX_BUCKET_MAX`): `(max_model_len - block_size) // block_size`
-    - Decode:
-      - batch size min (`VLLM_DECODE_BS_BUCKET_MIN`): `1`
-      - batch size step (`VLLM_DECODE_BS_BUCKET_STEP`): `min(max_num_seqs, 32)`
-      - batch size max (`VLLM_DECODE_BS_BUCKET_MAX`): `max_num_seqs`
-      - block size min (`VLLM_DECODE_BLOCK_BUCKET_MIN`): `block_size`
-      - block size step (`VLLM_DECODE_BLOCK_BUCKET_STEP`): `block_size`
-      - block size max (`VLLM_DECODE_BLOCK_BUCKET_MAX`): `max(128, (max_num_seqs*2048)/block_size)`
-  - Recommended Values:
-    - Prompt:
-      - sequence length max (`VLLM_PROMPT_SEQ_BUCKET_MAX`): `max_model_len`
-    - Decode:
-
-      - block size max (`VLLM_DECODE_BLOCK_BUCKET_MAX`): `max(128, (max_num_seqs*max_model_len/block_size)`
+- `VLLM_SKIP_WARMUP`: if `true`, warmup is skipped. The default is `false`.
 
 !!! note
     If the model config reports a high `max_model_len`, set it to max `input_tokens+output_tokens` rounded up to a multiple of `block_size` as per actual requirements.
@@ -69,3 +38,28 @@ Additionally, there are HPU PyTorch Bridge environment variables impacting vLLM
 - `PT_HPU_ENABLE_LAZY_COLLECTIVES`: must be set to `true` for tensor parallel inference with HPU Graphs. The default is `true`.
 - `PT_HPUGRAPH_DISABLE_TENSOR_CACHE`: must be set to `false` for LLaVA, qwen, and RoBERTa models. The default is `false`.
 - `VLLM_PROMPT_USE_FLEX_ATTENTION`: enabled only for the Llama model, allowing usage of `torch.nn.attention.flex_attention` instead of FusedSDPA. Requires `VLLM_PROMPT_USE_FUSEDSDPA=0`. The default is `false`.
+
+**Additional Performance Tuning Knobs - Linear Bucketing Strategy only:**
+
+- `VLLM_{phase}_{dim}_BUCKET_{param}`: collection of 12 environment variables configuring ranges of bucketing mechanism (linear bucketing only).
+  - `{phase}` is either `PROMPT` or `DECODE`
+  - `{dim}` is either `BS`, `SEQ` or `BLOCK`
+  - `{param}` is either `MIN`, `STEP` or `MAX`
+  - Default values:
+    - Prompt:
+      - batch size min (`VLLM_PROMPT_BS_BUCKET_MIN`): `1`
+      - batch size step (`VLLM_PROMPT_BS_BUCKET_STEP`): `32`
+      - batch size max (`VLLM_PROMPT_BS_BUCKET_MAX`): `max_num_prefill_seqs`
+      - sequence length min (`VLLM_PROMPT_SEQ_BUCKET_MIN`): `block_size`
+      - sequence length step (`VLLM_PROMPT_SEQ_BUCKET_STEP`): `block_size`
+      - sequence length max (`VLLM_PROMPT_SEQ_BUCKET_MAX`): `max_model_len`
+      - sequence ctx min (`VLLM_PROMPT_CTX_BUCKET_MIN`): `0`
+      - sequence ctx step (`VLLM_PROMPT_CTX_BUCKET_STEP`): `1`
+      - sequence ctx max (`VLLM_PROMPT_CTX_BUCKET_MAX`): `(max_model_len - block_size) // block_size`
+    - Decode:
+      - batch size min (`VLLM_DECODE_BS_BUCKET_MIN`): `1`
+      - batch size step (`VLLM_DECODE_BS_BUCKET_STEP`): `32`
+      - batch size max (`VLLM_DECODE_BS_BUCKET_MAX`): `max_num_seqs`
+      - block size min (`VLLM_DECODE_BLOCK_BUCKET_MIN`): `block_size`
+      - block size step (`VLLM_DECODE_BLOCK_BUCKET_STEP`): `block_size`
+      - block size max (`VLLM_DECODE_BLOCK_BUCKET_MAX`): `max_blocks`
diff --git a/docs/features/bucketing_mechanism.md b/docs/features/bucketing_mechanism.md
index 88d1f7d8..6a98cdc2 100644
--- a/docs/features/bucketing_mechanism.md
+++ b/docs/features/bucketing_mechanism.md
@@ -9,23 +9,69 @@ Intel Gaudi accelerators perform best when operating on models with fixed tensor
 generates optimized binary code that implements the given model topology on Gaudi. In its default configuration, the produced binary code may be highly dependent on input and output tensor shapes, requiring graph recompilation
 when encountering tensors with different shapes within the same topology. While these binaries efficiently utilize Gaudi, the compilation process itself can introduce noticeable overhead in end-to-end execution.
 In dynamic inference serving scenarios, minimizing the number of graph compilations and reducing the risk of graph compilation occurring during server runtime is important. Currently, this is achieved by
-"bucketing" the model's forward pass across two dimensions: `batch_size` and `sequence_length`.
+"bucketing" the model's forward pass across three dimensions.
 
-!!! note
-    Bucketing helps significantly reduce the number of required graphs, but does not handle graph compilation or device code generation. These tasks are performed during the warmup and HPUGraph capture phase.
+> [!NOTE]
+> Bucketing helps significantly reduce the number of required graphs, but does not handle graph compilation or device code generation. These tasks are performed during the warmup and HPUGraph capture phase.
+
+## Bucketing Strategies
+
+Bucketing is focused on three dimensions:
+- `batch size`: number of samples in batch
+- `query lenght`: sequence length without context tokens
+- `num blocks`: context length counted in blocks
 
-Bucketing ranges are determined with 3 parameters - `min`, `step`, and `max`. They can be set separately for the prompt and decode phase, and batch size and sequence length dimensions. These parameters
-can be observed in logs during vLLM startup:
+Bucketing ranges are generated based on 4 parameters - `min`, `step`, `max` and `limit`, separately for the prompt and decode phase, and batch size, query length and context blocks dimensions. These parameters can be observed in logs during vLLM startup:
 
 ```{.}
-INFO 08-01 21:37:59 hpu_model_runner.py:493] Prompt bucket config (min, step, max_warmup) bs:[1, 32, 4], seq:[128, 128, 1024]
-INFO 08-01 21:37:59 hpu_model_runner.py:499] Generated 24 prompt buckets: [(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024)]
-INFO 08-01 21:37:59 hpu_model_runner.py:504] Decode bucket config (min, step, max_warmup) bs:[1, 128, 4], seq:[128, 128, 2048]
-INFO 08-01 21:37:59 hpu_model_runner.py:509] Generated 48 decode buckets: [(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (1, 1152), (1, 1280), (1, 1408), (1, 1536), (1, 1664), (1, 1792), (1, 1920), (1, 2048), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (2, 1152), (2, 1280), (2, 1408), (2, 1536), (2, 1664), (2, 1792), (2, 1920), (2, 2048), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024), (4, 1152), (4, 1280), (4, 1408), (4, 1536), (4, 1664), (4, 1792), (4, 1920), (4, 2048)]
+INFO 07-07 19:27:37 [exponential.py:36] Prompt bucket config (min, step, max_warmup, limit) bs:[1, 1, 1, 1], seq:[128, 128, 1024, 11]
+INFO 07-07 19:27:37 [common.py:85] Generated 36 prompt buckets [bs, query, num_blocks]: [(1, 128, 0), (1, 128, 1), (1, 128, 2), (1, 128, 3), (1, 128, 4), (1, 128, 5), (1, 128, 6), (1, 128, 7), (1, 256, 0), (1, 256, 1), (1, 256, 2), (1, 256, 3), (1, 256, 4), (1, 256, 5), (1, 256, 6), (1, 384, 0), (1, 384, 1), (1, 384, 2), (1, 384, 3), (1, 384, 4), (1, 384, 5), (1, 512, 0), (1, 512, 1), (1, 512, 2), (1, 512, 3), (1, 512, 4), (1, 640, 0), (1, 640, 1), (1, 640, 2), (1, 640, 3), (1, 768, 0), (1, 768, 1), (1, 768, 2), (1, 896, 0), (1, 896, 1), (1, 1024, 0)]
+INFO 07-07 19:27:37 [common.py:85] Generated 42 decode buckets [bs, query, num_blocks]: [(1, 1, 128), (1, 1, 256), (1, 1, 384), (1, 1, 512), (1, 1, 640), (1, 1, 768), (1, 1, 896), (1, 1, 1024), (1, 1, 1408), (1, 1, 1792), (1, 1, 2432), (1, 1, 3328), (1, 1, 4352), (1, 1, 5888), (2, 1, 128), (2, 1, 256), (2, 1, 384), (2, 1, 512), (2, 1, 640), (2, 1, 768), (2, 1, 896), (2, 1, 1024), (2, 1, 1408), (2, 1, 1792), (2, 1, 2432), (2, 1, 3328), (2, 1, 4352), (2, 1, 5888), (4, 1, 128), (4, 1, 256), (4, 1, 384), (4, 1, 512), (4, 1, 640), (4, 1, 768), (4, 1, 896), (4, 1, 1024), (4, 1, 1408), (4, 1, 1792), (4, 1, 2432), (4, 1, 3328), (4, 1, 4352), (4, 1, 5888)]
 ```
 
-`min` determines the lowest value of the bucket. `step` determines the interval between buckets, and `max` determines the upper bound of the bucket. Furthermore, the interval between `min` and `step` has special handling - `min` gets multiplied by consecutive powers of two, until the multiplier is less than or equal to `step`. We call this the ramp-up phase, and it is used for handling lower batch sizes with minimum wastage,
-while allowing larger padding on larger batch sizes.
+> [!WARNING]
+> If a request exceeds the maximum bucket size in any dimension, it will be processed without padding, and its processing may require a graph compilation, potentially significantly increasing end-to-end latency.
+The boundaries of the buckets are user-configurable via environment variables, and upper bucket boundaries can be increased to avoid such scenario.
+
+For example, if a request with 3 sequences, each having a maximum sequence length of 412, is sent to an idle vLLM server, it will be padded and executed as a `(4, 512, 0)` prefill bucket, WHERE 4=bs, 512 .... This is because the `batch_size`
+(number of sequences) will be padded to 4 (the nearest batch size dimension higher than 3), and the maximum sequence length will be padded to 512 (the nearest sequence length dimension higher than 412). After the
+prefill stage, it will be executed as a `(4, 1, 512)` decode bucket and will remain in this bucket until either the batch dimension changes (e.g., due to a request being completed), in which case it will become
+a `(2, 1, 512)` bucket, or the context length increases beyond 512 tokens. It will become a `(4, 1, 640)` bucket at that point.
+
+> [!NOTE]
+> Bucketing is transparent to the user – padding in the sequence length dimension is never returned, and padding in the batch dimension does not create new requests.
+
+### Exponential Strategy  - Default
+
+Exponential strategy is the default warm-up mechanism. It is based on 4 parameters:
+- `min`: the smallest value
+- `step`: the rounding value for bucket boundaries
+- `max`: the largest value
+- `limit`: the maximum number of buckets
+
+> [!WARNING]
+> These parameters are not configurable by the user.
+
+The exponential bucketing strategy applies exponential spacing between buckets. The `min` and `max` values are always included in the warm-up, and the intermediate values are calculated using an exponent. The base remains unchanged. If duplicate values are generated, they are removed to ensure the warm-up process is as efficient as possible. All the values generated in this way, ranging from batch size, query length and context blocks, will be warmed up with each other.
+
+Example distribution is shown below:
+
+```{.}
+min = 128, step = 128, max = 4096, limit = 13
+```
+
+![exponential bucketing distribution for 4096 max query length](../../docs/assets/graphs/exponential_bucketing_example.png)
+
+This strategy creates more buckets with smaller values closer to `min`. As the values increase toward `max`, the buckets become less frequent, meaning the distance between them gets larger. This helps prioritize warming up the smaller values more precisely, while still covering the full range.
+
+### Linear Strategy
+
+> [!NOTE]
+> Starting from v1.22.0 Intel Gaudi Software release, Linear strategy is no longer the default warm-up mechanism.
+
+Linear strategy is determined with 3 parameters only - `min`, `step` and `max`. They can be set separately for the prompt and decode phase, and batch size and sequence length dimensions, by user.
+
+`min` determines the lowest value of the bucket. `step` determines the interval between buckets, and `max` determines the upper bound of the bucket. Furthermore, the interval between `min` and `step` has special handling: `min` is multiplied by consecutive powers of two until the multiplier is less than or equal to `step`. We refer to this as the ramp-up phase, which is used for handling lower batch sizes with minimal wastage, while allowing for larger padding on larger batch sizes.
 
 **Example with ramp-up**
 
@@ -45,115 +91,43 @@ min = 128, step = 128, max = 512
 => buckets = ramp_up + stable => (128, 256, 384, 512)
 ```
 
-In the logged scenario, 24 buckets were generated for prompt (prefill) runs, and 48 buckets for decode runs. Each bucket corresponds to a separate optimized device binary for a given model with specified tensor
-shapes. Whenever a batch of requests is processed, it is padded across batch and sequence length dimension to the smallest possible bucket.
+### Unified Strategy
 
-> [!WARNING]
-> If a request exceeds the maximum bucket size in any dimension, it will be processed without padding, and its processing may require a graph compilation, potentially significantly increasing end-to-end latency.
-The boundaries of the buckets are user-configurable via environment variables, and upper bucket boundaries can be increased to avoid such scenario.
+Unified strategy is dedicated strategy for Unified Attention. It's buckets are determined by different dimensions:
+- `query length`: number of currently processed tokens, without context tokens
+- `shared num blocks`: context length counted in blocks, including only blocks that are either shared between at least two block tables (different requests) or is used by at least two tokens in query
+- `unique num blocks`: context length counted in blocks, including only blocks that are not shared between block tables and are used only by one token
+- `is causal`: only two possible values: 0 and 1. Causal determines if there is at least one prompt in batch
 
-For example, if a request with 3 sequences, each having a maximum sequence length of 412, is sent to an idle vLLM server, it will be padded and executed as a `(4, 512)` prefill bucket. This is because the `batch_size`
-(number of sequences) will be padded to 4 (the nearest batch size dimension higher than 3), and the maximum sequence length will be padded to 512 (the nearest sequence length dimension higher than 412). After the
-prefill stage, it will be executed as a `(4, 512)` decode bucket and will remain in this bucket until either the batch dimension changes (e.g., due to a request being completed), in which case it will become
-a `(2, 512)` bucket, or the context length increases beyond 512 tokens. It will become a `(4, 640)` bucket at that point.
+Unified bucketing prepares buckets for both prompt and decode as one, known as `unified cfg`.
 
-> [!NOTE]
-> Bucketing is transparent to the user – padding in the sequence length dimension is never returned, and padding in the batch dimension does not create new requests.
+> [!WARNING]
+> No parameters for unified warmup are configurable by the user.
 
-## Warmup
+**Alpha Version:**
 
-Warmup is an optional but highly recommended step that occurs before the vLLM server starts listening. It executes a forward pass for each bucket using dummy data. The goal is to pre-compile all graphs
-and avoid any graph compilation overhead within bucket boundaries during server runtime. Each warmup step is logged during vLLM startup.
+Currently there are six points in ranges for query length, shared blocks and unique blocks. They are based on `max num seqs` and `max num batched tokens` values. Points are as follows whole, half and one quarter of both values resulting in six points in total.
 
-This example uses the same buckets as those in the Bucketing Mechanism section. Each output line corresponds to the execution of a single bucket. When a bucket is executed for the first time, its graph
-is compiled and can be reused later, avoiding further graph compilations.
+Example distribution is shown below:
 
 ```{.}
-INFO 08-01 22:26:47 hpu_model_runner.py:1066] [Warmup][Prompt][1/24] batch_size:4 seq_len:1024 free_mem:79.16 GiB
-INFO 08-01 22:26:47 hpu_model_runner.py:1066] [Warmup][Prompt][2/24] batch_size:4 seq_len:896 free_mem:55.43 GiB
-INFO 08-01 22:26:48 hpu_model_runner.py:1066] [Warmup][Prompt][3/24] batch_size:4 seq_len:768 free_mem:55.43 GiB
-...
-INFO 08-01 22:26:59 hpu_model_runner.py:1066] [Warmup][Prompt][24/24] batch_size:1 seq_len:128 free_mem:55.43 GiB
-INFO 08-01 22:27:00 hpu_model_runner.py:1066] [Warmup][Decode][1/48] batch_size:4 seq_len:2048 free_mem:55.43 GiB
-INFO 08-01 22:27:00 hpu_model_runner.py:1066] [Warmup][Decode][2/48] batch_size:4 seq_len:1920 free_mem:55.43 GiB
-INFO 08-01 22:27:01 hpu_model_runner.py:1066] [Warmup][Decode][3/48] batch_size:4 seq_len:1792 free_mem:55.43 GiB
-...
-INFO 08-01 22:27:16 hpu_model_runner.py:1066] [Warmup][Decode][47/48] batch_size:2 seq_len:128 free_mem:55.43 GiB
-INFO 08-01 22:27:16 hpu_model_runner.py:1066] [Warmup][Decode][48/48] batch_size:1 seq_len:128 free_mem:55.43 GiB
+batch size = 64, max num batched tokens = 4096
 ```
 
-> [!TIP]
-> Compiling all the buckets may take some time and can be disabled by setting the `VLLM_SKIP_WARMUP=true` environment variable. Remember that if you do this, you may encounter graph compilations
-when executing a given bucket for the first time.
+![exponential bucketing distribution for 4096 max query length](../../docs/assets/graphs/unified_bucketing_example.png)
 
-> [!WARNING]
-> Disabling warmup is fine for development, but it is highly recommended to enable it in deployment.
-
-## HPU Graph Capture
-
-[HPU Graphs](https://docs.habana.ai/en/latest/PyTorch/Inference_on_PyTorch/Inference_Using_HPU_Graphs.html) are currently the most performant execution method of vLLM on Intel Gaudi. When HPU Graphs are enabled,
-execution graphs will be traced (recorded) ahead of time (after performing warmup), to be later replayed during inference, significantly reducing host overheads. Recording can take large amounts of memory, which
-needs to be taken into account when allocating KV cache. Enabling HPU Graphs will impact the number of available KV cache blocks, but vLLM provides user-configurable variables to control memory management.
-
-When HPU Graphs are used, they share the common memory pool ("usable memory") with the KV cache, as determined by the `gpu_memory_utilization` flag (default value is `0.9`). Before the KV cache is allocated,
-the model weights are loaded onto the device, and a forward pass of the model is executed on dummy data to estimate memory usage. Only after that, the `gpu_memory_utilization` flag is applied. At its default value,
-it marks 90% of the free device memory at that point as usable. Next, the KV cache is allocated, the model is warmed up, and HPU Graphs are captured. The `VLLM_GRAPH_RESERVED_MEM` environment variable defines
-the ratio of memory reserved for HPU Graph capture. With its default value (`VLLM_GRAPH_RESERVED_MEM=0.1`), 10% of the usable memory will be reserved for graph capture (referred to as "usable graph memory"),
-and the remaining 90% will be used for the KV cache. The environment variable `VLLM_GRAPH_PROMPT_RATIO` determines the ratio of usable graph memory reserved for prefill and decode graphs. A lower value corresponds to less usable graph memory reserved for the prefill stage. For example, setting `VLLM_GRAPH_PROMPT_RATIO=0.2`
-reserves 20% of usable graph memory for prefill graphs, while 80% is allocated for decode graphs.
-
-> [!NOTE]
-> `gpu_memory_utilization` does not represent the absolute memory usage across the HPU. Instead, it specifies the memory margin after loading the model and running a profile. For example, if a device has 100 GiB of
-total memory and 50 GiB of free memory after loading the model weights and executing the profiling run, the default value of `gpu_memory_utilization` will mark 90% of the 50 GiB as usable, leaving 5 GiB as a margin,
-regardless of the total device memory.
+Additionaly for context blocks, both shared and unique, `0` value will be added as well.
 
-You can also configure the strategy for capturing HPU graphs separately for the prompt and decode stages. The strategy affects the order in which graphs are captured. Two strategies are implemented:
+This way our bucketing will look like this:
 
-- `max_bs` - The graph capture queue is sorted in descending order by batch size. Buckets with equal batch sizes are sorted by sequence length in ascending order
-  (e.g., `(64, 128)`, `(64, 256)`, `(32, 128)`, `(32, 256)`, `(1, 128)`, `(1,256)`), which is the default strategy for decode.
-- `min_tokens` - The graph capture queue is sorted in ascending order by the number of tokens each graph processes (`batch_size*sequence_length`), which is the default strategy for prompt.
-
-When many requests are pending, the vLLM scheduler attempts to fill the maximum batch size for decoding as quickly as possible. Once a request is finished, the decode batch size decreases.
-When this happens, vLLM attempts to schedule a prefill iteration for requests in the waiting queue to restore the decode batch size to its previous state. In a fully loaded scenario, the decode
-batch size is often at its maximum, making large-batch HPU graphs critical to capture, as indicated by the `max_bs` strategy. Conversely, prefill iterations will typically be executed with very low
-batch sizes (1-4), as reflected in the `min_tokens` strategy.
-
-> [!NOTE]
-> `VLLM_GRAPH_PROMPT_RATIO` does not set a hard limit on the memory allocated for graphs in each stage (prefill and decode). vLLM first attempts to use the entire usable prefill graph memory
-(usable graph memory * VLLM_GRAPH_PROMPT_RATIO) to capture prefilled HPU graphs. It will then attempt to do the same for decode graphs and the usable decode graph memory pool. If one stage is fully
-captured and there is unused memory remaining in the usable graph memory pool, vLLM will attempt to capture more graphs for the other stage, until no more HPU Graphs can be captured without exceeding
-the reserved memory pool. The behavior of this mechanism is illustrated in the example below.
+```{.}
+INFO 09-23 12:32:43 [common.py:100] Generated 375 unified buckets [query, shared_blocks, unique_blocks]: [(8, 0, 0, 1), (8, 0, 8, 0), ..., (2048, 256, 2890, 1), (2048, 256, 5781, 1)]
+```
 
-Each step outlined is logged by the vLLM server, with negative values indicating memory release:
+With every bucket logged separately in warm-up phase:
 
 ```{.}
-INFO 08-02 17:37:44 hpu_model_runner.py:493] Prompt bucket config (min, step, max_warmup) bs:[1, 32, 4], seq:[128, 128, 1024]
-INFO 08-02 17:37:44 hpu_model_runner.py:499] Generated 24 prompt buckets: [(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024)]
-INFO 08-02 17:37:44 hpu_model_runner.py:504] Decode bucket config (min, step, max_warmup) bs:[1, 128, 4], seq:[128, 128, 2048]
-INFO 08-02 17:37:44 hpu_model_runner.py:509] Generated 48 decode buckets: [(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (1, 1152), (1, 1280), (1, 1408), (1, 1536), (1, 1664), (1, 1792), (1, 1920), (1, 2048), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (2, 1152), (2, 1280), (2, 1408), (2, 1536), (2, 1664), (2, 1792), (2, 1920), (2, 2048), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024), (4, 1152), (4, 1280), (4, 1408), (4, 1536), (4, 1664), (4, 1792), (4, 1920), (4, 2048)]
-INFO 08-02 17:37:52 hpu_model_runner.py:430] Pre-loading model weights on hpu:0 took 14.97 GiB of device memory (14.97 GiB/94.62 GiB used) and 2.95 GiB of host memory (475.2 GiB/1007 GiB used)
-INFO 08-02 17:37:52 hpu_model_runner.py:438] Wrapping in HPU Graph took 0 B of device memory (14.97 GiB/94.62 GiB used) and -252 KiB of host memory (475.2 GiB/1007 GiB used)
-INFO 08-02 17:37:52 hpu_model_runner.py:442] Loading model weights took in total 14.97 GiB of device memory (14.97 GiB/94.62 GiB used) and 2.95 GiB of host memory (475.2 GiB/1007 GiB used)
-INFO 08-02 17:37:54 hpu_worker.py:134] Model profiling run took 504 MiB of device memory (15.46 GiB/94.62 GiB used) and 180.9 MiB of host memory (475.4 GiB/1007 GiB used)
-INFO 08-02 17:37:54 hpu_worker.py:158] Free device memory: 79.16 GiB, 39.58 GiB usable (gpu_memory_utilization=0.5), 15.83 GiB reserved for HPUGraphs (VLLM_GRAPH_RESERVED_MEM=0.4), 23.75 GiB reserved for KV cache
-INFO 08-02 17:37:54 hpu_executor.py:85] # HPU blocks: 1519, # CPU blocks: 0
-INFO 08-02 17:37:54 hpu_worker.py:190] Initializing cache engine took 23.73 GiB of device memory (39.2 GiB/94.62 GiB used) and -1.238 MiB of host memory (475.4 GiB/1007 GiB used)
-INFO 08-02 17:37:54 hpu_model_runner.py:1066] [Warmup][Prompt][1/24] batch_size:4 seq_len:1024 free_mem:55.43 GiB
-...
-INFO 08-02 17:38:22 hpu_model_runner.py:1066] [Warmup][Decode][48/48] batch_size:1 seq_len:128 free_mem:55.43 GiB
-INFO 08-02 17:38:22 hpu_model_runner.py:1159] Using 15.85 GiB/55.43 GiB of free device memory for HPUGraphs, 4.755 GiB for prompt and 11.095 GiB for decode (VLLM_GRAPH_PROMPT_RATIO=0.3)
-INFO 08-02 17:38:22 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][1/24] batch_size:1 seq_len:128 free_mem:55.43 GiB
-...
-INFO 08-02 17:38:26 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][11/24] batch_size:1 seq_len:896 free_mem:48.77 GiB
-INFO 08-02 17:38:27 hpu_model_runner.py:1066] [Warmup][Graph/Decode][1/48] batch_size:4 seq_len:128 free_mem:47.51 GiB
-...
-INFO 08-02 17:38:41 hpu_model_runner.py:1066] [Warmup][Graph/Decode][48/48] batch_size:1 seq_len:2048 free_mem:47.35 GiB
-INFO 08-02 17:38:41 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][12/24] batch_size:4 seq_len:256 free_mem:47.35 GiB
-INFO 08-02 17:38:42 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][13/24] batch_size:2 seq_len:512 free_mem:45.91 GiB
-INFO 08-02 17:38:42 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][14/24] batch_size:1 seq_len:1024 free_mem:44.48 GiB
-INFO 08-02 17:38:43 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][15/24] batch_size:2 seq_len:640 free_mem:43.03 GiB
-INFO 08-02 17:38:43 hpu_model_runner.py:1128] Graph/Prompt captured:15 (62.5%) used_mem:14.03 GiB buckets:[(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (4, 128), (4, 256)]
-INFO 08-02 17:38:43 hpu_model_runner.py:1128] Graph/Decode captured:48 (100.0%) used_mem:161.9 MiB buckets:[(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (1, 1152), (1, 1280), (1, 1408), (1, 1536), (1, 1664), (1, 1792), (1, 1920), (1, 2048), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (2, 1152), (2, 1280), (2, 1408), (2, 1536), (2, 1664), (2, 1792), (2, 1920), (2, 2048), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024), (4, 1152), (4, 1280), (4, 1408), (4, 1536), (4, 1664), (4, 1792), (4, 1920), (4, 2048)]
-INFO 08-02 17:38:43 hpu_model_runner.py:1206] Warmup finished in 49 secs, allocated 14.19 GiB of device memory
-INFO 08-02 17:38:43 hpu_executor.py:91] init_cache_engine took 37.92 GiB of device memory (53.39 GiB/94.62 GiB used) and 57.86 MiB of host memory (475.4 GiB/1007 GiB used)
+(EngineCore_DP0 pid=805) INFO 09-23 12:32:50 [hpu_model_runner.py:3320] [Warmup][Unified CFG][2/375] query_len:2048 shared_blocks:256 unique_blocks:2890 (causal) free_mem:11.16 GiB
+(EngineCore_DP0 pid=805) INFO 09-23 12:32:53 [hpu_model_runner.py:3320] [Warmup][Unified CFG][3/375] query_len:2048 shared_blocks:256 unique_blocks:1445 (causal) free_mem:11.16 GiB
+(EngineCore_DP0 pid=805) INFO 09-23 12:32:56 [hpu_model_runner.py:3320] [Warmup][Unified CFG][4/375] query_len:2048 shared_blocks:256 unique_blocks:32 (causal) free_mem:11.16 GiB
 ```
diff --git a/docs/features/warmup.md b/docs/features/warmup.md
new file mode 100644
index 00000000..a64d2582
--- /dev/null
+++ b/docs/features/warmup.md
@@ -0,0 +1,225 @@
+---
+title: Warm-up
+---
+[](){ #Warm-up }
+
+## Warm-up
+
+Warm-up is a highly recommended step that occurs before the vLLM server starts listening. It performs a forward pass for each bucket using dummy data. The goal is to precompile all graphs and eliminate any graph compilation overhead within bucket boundaries during server runtime. Each warmup step is logged during vLLM startup.
+This example uses the same buckets as those described in the Bucketing Mechanism section. Each output line corresponds to the execution of a single bucket. When a bucket is executed for the first time, its graph is compiled and can be reused later, avoiding further graph compilations.
+
+```{.}
+INFO 08-01 22:26:47 hpu_model_runner.py:1066] [Warmup][Prompt][1/24] batch_size:4 seq_len:1024 free_mem:79.16 GiB
+INFO 08-01 22:26:47 hpu_model_runner.py:1066] [Warmup][Prompt][2/24] batch_size:4 seq_len:896 free_mem:55.43 GiB
+INFO 08-01 22:26:48 hpu_model_runner.py:1066] [Warmup][Prompt][3/24] batch_size:4 seq_len:768 free_mem:55.43 GiB
+...
+INFO 08-01 22:26:59 hpu_model_runner.py:1066] [Warmup][Prompt][24/24] batch_size:1 seq_len:128 free_mem:55.43 GiB
+INFO 08-01 22:27:00 hpu_model_runner.py:1066] [Warmup][Decode][1/48] batch_size:4 seq_len:2048 free_mem:55.43 GiB
+INFO 08-01 22:27:00 hpu_model_runner.py:1066] [Warmup][Decode][2/48] batch_size:4 seq_len:1920 free_mem:55.43 GiB
+INFO 08-01 22:27:01 hpu_model_runner.py:1066] [Warmup][Decode][3/48] batch_size:4 seq_len:1792 free_mem:55.43 GiB
+...
+INFO 08-01 22:27:16 hpu_model_runner.py:1066] [Warmup][Decode][47/48] batch_size:2 seq_len:128 free_mem:55.43 GiB
+INFO 08-01 22:27:16 hpu_model_runner.py:1066] [Warmup][Decode][48/48] batch_size:1 seq_len:128 free_mem:55.43 GiB
+```
+
+> [!TIP]
+> Compiling all buckets may take some time and can be disabled by setting the `VLLM_SKIP_WARMUP=true` environment variable. Keep in mind that if you do this, you may encounter graph compilations when executing a given bucket for the first time.
+
+> [!WARNING]
+> Disabling warmup is acceptable for development purposes, but it is strongly recommended to enable it in production deployments.
+
+## HPU Graph Capture
+
+[HPU Graphs](https://docs.habana.ai/en/latest/PyTorch/Inference_on_PyTorch/Inference_Using_HPU_Graphs.html) are currently the most performant execution method for vLLM on Intel Gaudi. When HPU Graphs are enabled,
+execution graphs are be traced (recorded) ahead of time (after performing warmup), and later replayed during inference, significantly reducing host overheads. Recording can consume large amounts of memory, which
+must be considred when allocating KV cache. Enabling HPU Graphs affects the number of available KV cache blocks, but vLLM provides user-configurable variables to manage memory usage.
+
+When HPU Graphs are used, they share the common memory pool ("usable memory") with the KV cache, as determined by the `gpu_memory_utilization` flag (default value is `0.9`). Before allocating the KV cache,
+the model weights are loaded onto the device, and a forward pass is executed on dummy data to estimate memory usage. Only then is the `gpu_memory_utilization` flag applied. At its default value,
+it marks 90% of the free device memory at that point as usable. Next, the KV cache is allocated, the model is warmed up, and HPU Graphs are captured. The `VLLM_GRAPH_RESERVED_MEM` environment variable defines
+the ratio of memory reserved for HPU Graph capture. With its default value (`VLLM_GRAPH_RESERVED_MEM=0.1`), 10% of the usable memory will be reserved for graph capture (referred to as "usable graph memory"), and the remaining 90% will be used for the KV cache.
+
+> [!NOTE]
+> `gpu_memory_utilization` does not represent the absolute memory usage across the HPU. Instead, it specifies the memory margin after loading the model and running a profiling pass. For example, if a device has 100 GiB of total memory and 50 GiB of free memory after loading the model weights and executing the profiling run, the default value of `gpu_memory_utilization` will mark 90% of the 50 GiB as usable, leaving 5 GiB as a margin - regardless of the total device memory.
+
+When many requests are pending, the vLLM scheduler attempts to fill the maximum decode batch size as quickly as possible. Once a request completes, the decode batch size decreases. When this happens, vLLM schedules a prefill iteration for requests in the waiting queue to restore the previous decode batch size. In fully loaded scenarios, the decode batch size is often at its maximum, making large-batch HPU graphs critical to capture. On the other hand, prompt iterations typically execute with very low batch sizes (1-4).
+
+Each step outlined is logged by the vLLM server, with negative values indicating memory release:
+
+```{.}
+INFO 08-02 17:37:44 hpu_model_runner.py:493] Prompt bucket config (min, step, max_warmup) bs:[1, 32, 4], seq:[128, 128, 1024]
+INFO 08-02 17:37:44 hpu_model_runner.py:499] Generated 24 prompt buckets: [(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024)]
+INFO 08-02 17:37:44 hpu_model_runner.py:504] Decode bucket config (min, step, max_warmup) bs:[1, 128, 4], seq:[128, 128, 2048]
+INFO 08-02 17:37:44 hpu_model_runner.py:509] Generated 48 decode buckets: [(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (1, 1152), (1, 1280), (1, 1408), (1, 1536), (1, 1664), (1, 1792), (1, 1920), (1, 2048), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (2, 1152), (2, 1280), (2, 1408), (2, 1536), (2, 1664), (2, 1792), (2, 1920), (2, 2048), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024), (4, 1152), (4, 1280), (4, 1408), (4, 1536), (4, 1664), (4, 1792), (4, 1920), (4, 2048)]
+INFO 08-02 17:37:52 hpu_model_runner.py:430] Pre-loading model weights on hpu:0 took 14.97 GiB of device memory (14.97 GiB/94.62 GiB used) and 2.95 GiB of host memory (475.2 GiB/1007 GiB used)
+INFO 08-02 17:37:52 hpu_model_runner.py:438] Wrapping in HPU Graph took 0 B of device memory (14.97 GiB/94.62 GiB used) and -252 KiB of host memory (475.2 GiB/1007 GiB used)
+INFO 08-02 17:37:52 hpu_model_runner.py:442] Loading model weights took in total 14.97 GiB of device memory (14.97 GiB/94.62 GiB used) and 2.95 GiB of host memory (475.2 GiB/1007 GiB used)
+INFO 08-02 17:37:54 hpu_worker.py:134] Model profiling run took 504 MiB of device memory (15.46 GiB/94.62 GiB used) and 180.9 MiB of host memory (475.4 GiB/1007 GiB used)
+INFO 08-02 17:37:54 hpu_worker.py:158] Free device memory: 79.16 GiB, 39.58 GiB usable (gpu_memory_utilization=0.5), 15.83 GiB reserved for HPUGraphs (VLLM_GRAPH_RESERVED_MEM=0.4), 23.75 GiB reserved for KV cache
+INFO 08-02 17:37:54 hpu_executor.py:85] # HPU blocks: 1519, # CPU blocks: 0
+INFO 08-02 17:37:54 hpu_worker.py:190] Initializing cache engine took 23.73 GiB of device memory (39.2 GiB/94.62 GiB used) and -1.238 MiB of host memory (475.4 GiB/1007 GiB used)
+INFO 08-02 17:37:54 hpu_model_runner.py:1066] [Warmup][Prompt][1/24] batch_size:4 seq_len:1024 free_mem:55.43 GiB
+...
+INFO 08-02 17:38:22 hpu_model_runner.py:1066] [Warmup][Decode][48/48] batch_size:1 seq_len:128 free_mem:55.43 GiB
+INFO 08-02 17:38:22 hpu_model_runner.py:1159] Using 15.85 GiB/55.43 GiB of free device memory for HPUGraphs, 4.755 GiB for prompt and 11.095 GiB for decode (VLLM_GRAPH_PROMPT_RATIO=0.3)
+INFO 08-02 17:38:22 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][1/24] batch_size:1 seq_len:128 free_mem:55.43 GiB
+...
+INFO 08-02 17:38:26 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][11/24] batch_size:1 seq_len:896 free_mem:48.77 GiB
+INFO 08-02 17:38:27 hpu_model_runner.py:1066] [Warmup][Graph/Decode][1/48] batch_size:4 seq_len:128 free_mem:47.51 GiB
+...
+INFO 08-02 17:38:41 hpu_model_runner.py:1066] [Warmup][Graph/Decode][48/48] batch_size:1 seq_len:2048 free_mem:47.35 GiB
+INFO 08-02 17:38:41 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][12/24] batch_size:4 seq_len:256 free_mem:47.35 GiB
+INFO 08-02 17:38:42 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][13/24] batch_size:2 seq_len:512 free_mem:45.91 GiB
+INFO 08-02 17:38:42 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][14/24] batch_size:1 seq_len:1024 free_mem:44.48 GiB
+INFO 08-02 17:38:43 hpu_model_runner.py:1066] [Warmup][Graph/Prompt][15/24] batch_size:2 seq_len:640 free_mem:43.03 GiB
+INFO 08-02 17:38:43 hpu_model_runner.py:1128] Graph/Prompt captured:15 (62.5%) used_mem:14.03 GiB buckets:[(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (4, 128), (4, 256)]
+INFO 08-02 17:38:43 hpu_model_runner.py:1128] Graph/Decode captured:48 (100.0%) used_mem:161.9 MiB buckets:[(1, 128), (1, 256), (1, 384), (1, 512), (1, 640), (1, 768), (1, 896), (1, 1024), (1, 1152), (1, 1280), (1, 1408), (1, 1536), (1, 1664), (1, 1792), (1, 1920), (1, 2048), (2, 128), (2, 256), (2, 384), (2, 512), (2, 640), (2, 768), (2, 896), (2, 1024), (2, 1152), (2, 1280), (2, 1408), (2, 1536), (2, 1664), (2, 1792), (2, 1920), (2, 2048), (4, 128), (4, 256), (4, 384), (4, 512), (4, 640), (4, 768), (4, 896), (4, 1024), (4, 1152), (4, 1280), (4, 1408), (4, 1536), (4, 1664), (4, 1792), (4, 1920), (4, 2048)]
+INFO 08-02 17:38:43 hpu_model_runner.py:1206] Warmup finished in 49 secs, allocated 14.19 GiB of device memory
+INFO 08-02 17:38:43 hpu_executor.py:91] init_cache_engine took 37.92 GiB of device memory (53.39 GiB/94.62 GiB used) and 57.86 MiB of host memory (475.4 GiB/1007 GiB used)
+```
+
+## Sampler Warm-Up
+The sampler converts model logits into next-token selections, using configured decoding strategies (greedy or probabilistic). Its warm-up phase prepares compiled graph variants (or internal code paths) for a representative set of batch sizes and sampling parameter combinations, so that first real user requests avoid extra compilation/setup latency.
+
+### How the Sampler Warm-Up Works
+
+Implemented in `warmup_sampler`, the routine systematically exercises the sampling stack across a Cartesian set of (batch size, temperature, top-p, top-k) patterns and a flag, that signals whether the batch size changed. Key steps:
+
+1. Build a list of test batch sizes: it prepends `[0, 1]` to the distinct decode bucket batch sizes, as these need to be always warmed up.
+2. Define a list of sampling configurations (12 total) covering:
+   * Greedy (temperature=0.0)
+   * Typical random sampling (temperature=1.0)
+   * Creative settings (0.7/0.9/top-k=50)
+   * Conservative (0.3/0.95/top-k=20)
+   * High temperature (1.2/0.8/top-k=100)
+   * Top-p only variants (e.g. 0.8/0.85/top-k=0)
+    Each appears twice: once with `batch_changed=True` and once with `batch_changed=False` to exercise any internal fast-path or cache invalidation logic tied to batch resizing.
+3. For every batch size:
+   * Create a dummy hidden state tensor shaped `(batch_size, hidden_size)` and compute logits via `model.compute_logits`.
+   * Instantiate dummy request objects (at least one) with placeholder prompt tokens and single KV block.
+4. For each sampling configuration:
+   * Update each request's `SamplingParams` (temperature, top_p, top_k).
+   * Mark the request as greedy or random (separate sets) to test branching.
+   * Populate `req_output_token_ids` with padded placeholders and refresh internal sampling metadata.
+   * Invoke `_run_sampling` passing `batch_changed` so both changed/unchanged batch-size code paths get compiled/exercised.
+   * Reset per-iteration sampler bookkeeping sets/lists.
+5. After finishing all sampling configs for a batch size, clear request maps and continue.
+6. Perform an HPU synchronize and log success.
+
+### What the Logs Look Like
+
+Typical sequence:
+
+```text
+INFO 09-22 16:39:42 [hpu_model_runner.py:3347] Warming up sampler with batch sizes: [0, 1, 138] and following configs:
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.0, top_p=1.0, top_k=0, batch_changed=True
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=1.0, top_p=1.0, top_k=0, batch_changed=True
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.7, top_p=0.9, top_k=50, batch_changed=True
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.3, top_p=0.95, top_k=20, batch_changed=True
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=1.2, top_p=0.8, top_k=100, batch_changed=True
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.8, top_p=0.85, top_k=0, batch_changed=True
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.0, top_p=1.0, top_k=0, batch_changed=False
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=1.0, top_p=1.0, top_k=0, batch_changed=False
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.7, top_p=0.9, top_k=50, batch_changed=False
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.3, top_p=0.95, top_k=20, batch_changed=False
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=1.2, top_p=0.8, top_k=100, batch_changed=False
+INFO 09-22 16:39:42 [hpu_model_runner.py:3349] temp=0.8, top_p=0.85, top_k=0, batch_changed=False
+INFO 09-22 16:39:42 [hpu_model_runner.py:3350] Starting sampler warmup...
+INFO 09-22 16:39:43 [hpu_model_runner.py:3411] Sampler warmup completed successfully
+```
+
+If warm-up is globally skipped ([see below](#how-to-turn-it-off)), none of these lines appear.
+
+### Why We Warm Up the Sampler (and Risks If We Do Not)
+
+Without sampler warm-up:
+* The first real request using a new combination (e.g., first high-temp + top-k path, or first batch size after scaling up load) might incur graph recompilation, adding latency to that user request.
+* Tail latency variance increases: early heterogeneous workloads cause multiple staggered compilations.
+* Batch-size transition logic (paths where `batch_changed=True`) may pay initialization cost during live traffic.
+
+With warm-up:
+* Common sampling hyperparameter mixes are compiled ahead-of-time.
+* Greedy vs random branching and metadata refresh code paths are stabilized.
+* Batch growth/shrink handling is already exercised, smoothing later scaling behavior.
+
+Skipping the sampler warm-up does not affect correctness—only the latency profile of the earliest varied sampling requests.
+
+### How to Turn It Off
+
+There is no dedicated flag for the sampler alone. It participates in the global warm-up sequence and is skipped when:
+
+* `VLLM_SKIP_WARMUP=true` is set.
+* The engine is configured to enforce eager execution in a mode where no graph capture/compilation is desired (sampler still runs the first time on demand, but without a separate warm-up call).
+
+### Related Notes & Environment Variables
+
+* `VLLM_SKIP_WARMUP` – Disables sampler warm-up along with other warm-up phases.
+* Decode bucket configuration env vars indirectly influence the set of batch sizes the sampler warms up (since it derives test batch sizes from decode buckets).
+
+> [!NOTE]
+> If you introduce new sampling behaviors (e.g., new nucleus filtering, penalties, or speculative metadata), extend `sampling_configs` in `warmup_sampler` so their graph paths are primed.
+
+## Defragmenter Warm-Up
+
+The defragmenter reclaims and compacts sparse KV-cache block usage at runtime by swapping rarely packed high-index blocks with lower free indices. Its warm-up phase pre-compiles the small swap graphs so that later online defragmentation can execute with near-zero graph compile latency.
+
+### How the Defragmenter Warm-Up Works
+
+During the main warm-up (`warmup_model`) we call an internal method (`warmup_defragmenter`) after the KV caches and defragmenter have been initialized. The routine:
+
+1. Verifies the feature is enabled (defragmenter only runs when unified attention is enabled) and that swap utilities (`cache_utils`) are prepared.
+2. Determines the list of padding thresholds: `[8, 16, 32, 64, 128, 256, 512]`.
+3. Chooses a minimal valid swap pair `[(1, 0)]` (two distinct block IDs). Only two real blocks are required; internally each swap call is padded up to the current threshold length so that a compiled graph for that exact padded size is produced.
+4. Iterates through each threshold and invokes a swap. This captures/compiles (depending on execution mode) the swap graph for that padded size.
+5. If the number of thresholds is odd, performs one extra swap with the first threshold so that the sequence of swaps returns the KV cache to its original state (net zero logical change).
+6. Logs completion.
+
+Because every future real defragmentation swap request will round/pad to one of these known thresholds, all operational swap sizes hit a pre-compiled path and avoid on-demand compilation latency.
+
+### What the Logs Look Like
+
+You will typically see one of two flows. If there are at least two KV-cache blocks available:
+
+```text
+INFO 09-22 16:26:24 [hpu_model_runner.py:3428] Warming up defragmenter with thresholds: [8, 16, 32, 64, 128, 256, 512]
+INFO 09-22 16:26:27 [hpu_model_runner.py:3452] Defragmenter warmup completed successfully
+```
+
+If insufficient blocks exist (e.g., extremely small test configuration or allocation failure) warm-up is skipped gracefully:
+
+```text
+INFO 09-22 16:26:24 [hpu_model_runner.py:3428] Warming up defragmenter with thresholds: [8, 16, 32, 64, 128, 256, 512]
+WARNING hh:mm:ss hpu_model_runner.py:#### Skipping defragmenter warmup, insufficient blocks (1)
+```
+
+Add `VLLM_DEBUG=defrag` to the environment to emit fine-grained debug messages during live defragmentation (not during the minimal warm-up swaps only) such as the number of blocks swapped and post-compaction statistics.
+
+### Why We Warm Up (and What Happens If We Do Not)
+
+Defragmentation may be triggered mid-serving when the highest allocated block index drifts far above the actual number of in-use blocks (fragmentation). The operation itself is a sequence of swap kernels over key & value caches. Without warm-up:
+
+* The first fragmentation event that requires a new (previously unseen) padded swap size would incur graph capture/compilation in the critical path.
+* That added latency can surface as a sudden tail-latency spike for a user request.
+* Multiple different first-seen swap sizes across processes could each trigger separate compilations.
+
+With warm-up, all representative padded sizes are compiled ahead-of-time via a deterministic, tiny swap, so online defragmentation becomes a predictable, low-latency maintenance task.
+
+Skipping only the defragmenter warm-up does not break correctness; it only risks sporadic latency when fragmentation first crosses a threshold that mandates compaction.
+
+### How to Turn It Off
+
+You can disable (a) the warm-up step itself or (b) the entire defragmentation feature:
+
+* Disable all warm-up phases (including defragmenter) by setting `VLLM_SKIP_WARMUP=true`.
+* Run without unified attention (the defragmenter is tied to unified attention; if unified attention is disabled, `defrag` is not enabled and the warm-up is a no-op). There is no separate dedicated environment flag to force-enable/disable defrag beyond unified attention in this version.
+* Avoid graph compilation for defragmenter swaps by setting `VLLM_DEFRAG_WITH_GRAPHS=false` (falls back to regular execution; warm-up will still exercise swaps but without graph capture), if supported by the execution mode.
+
+Related environment variables:
+
+* `VLLM_DEFRAG_THRESHOLD` – Fragmentation trigger heuristic (default 32). Lower values make compaction more aggressive.
+* `VLLM_DEFRAG_WITH_GRAPHS` – Whether swap paths are compiled/graphed (defaults to `bridge_mode == eager`).
+* `VLLM_DEBUG=defrag` – Enables verbose defragmentation debug logging.
+* `VLLM_SKIP_WARMUP` – Disables all warm-up stages including this one.
+
+> [!NOTE]
+> Disabling defragmenter warm-up does not disable defragmentation itself (unless unified attention/the feature is off). It only removes ahead-of-time graph preparation, potentially pushing compile cost into the first live fragmentation event.