feat: use local_into_box

[giladchase]

• usage subject to dedicated db-flag
• after tracking local variables in analyze_ap_changes, into_box sierra-gen conditionally emits
  either into_box or local_into_box if the input is small enough to be worth it.

[reviewable-StarkWare]

This change is 

[giladchase]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• feat: use local_into_box #9046 👈 (View in Graphite)
• feat: add Unbox for symmetry with IntoBox #9080
• Add IntoBox instead of into_box in lowering #8967
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[orizi]

@orizi reviewed 2 of 10 files at r1, all commit messages.
Reviewable status: 2 of 10 files reviewed, 2 unresolved discussions (waiting on @eytan-starkware and @giladchase)

---

crates/cairo-lang-filesystem/src/flag.rs line 26 at r1 (raw file):

    ///
    /// Default is true.
    LocalIntoBoxOptimization(bool),

use "sierra_future" flag - should exist after you rebase.

so revert the file.

---

crates/cairo-lang-sierra-generator/src/local_variables.rs line 41 at r1 (raw file):

    pub ap_tracking_configuration: ApTrackingConfiguration,
    /// Variables that are known to be non-AP-based (expanded to include all aliases).
    pub non_ap_based_variables: UnorderedHashSet<VariableId>,

why isn't this just included in the set of local variables?

[orizi]

Reviewable status: 2 of 10 files reviewed, 3 unresolved discussions (waiting on @eytan-starkware and @giladchase)

---

crates/cairo-lang-sierra-generator/src/function_generator_test_data/boxing line 110 at r1 (raw file):

fn foo() -> Box<MyStruct> {
    let x = create_struct();
    revoke_ap();

why is the revoke required? it should be store_local anyway because it is large - no?

[orizi]

@orizi reviewed all commit messages.
Reviewable status: 0 of 10 files reviewed, 3 unresolved discussions (waiting on @eytan-starkware and @giladchase)