feat: Weighted cutflow

[NJManganelli]

This PR adds an (opinionated) take on adding weighted cutflows to coffea, permitting the cutflow to be instantiated with a Weights object and a weights modifier (i.e. a weights variation name, as in Weights.weights("jesUp"))

• Because of need for TypeError: meta must be an instance of an Awkward Array, not <class 'module'> triggered by fill_flattened for a dask-histogram and dask-awkward arrays dask-contrib/dask-awkward#531 need to roll-back to manual flattening of variables in plot_vars for now
• 4 tests currently broken because of above (and potentially other bugs)
• need expansion of tests to cover weighted fills
• Equivalent in NMinusOne would also be desirable, may want separate PR
• Docs showing usage, including commonmask argument

[lgray]

FAILED tests/test_analysis_tools.py::test_packed_selection_cutflow_dak[True] - TypeError: meta must be an instance of an Awkward Array, not <class 'module'>.
FAILED tests/test_analysis_tools.py::test_packed_selection_cutflow_dak[False] - TypeError: meta must be an instance of an Awkward Array, not <class 'module'>.
FAILED tests/test_analysis_tools.py::test_packed_selection_cutflow_dak_uproot_only[True] - TypeError: meta must be an instance of an Awkward Array, not <class 'module'>.
FAILED tests/test_analysis_tools.py::test_packed_selection_cutflow_dak_uproot_only[False] - TypeError: meta must be an instance of an Awkward Array, not <class 'module'>.

[NJManganelli]

Aye, those failures are the attempt to use fill_flattened in dask which doesn't currently work, so it will be rolled-back to manual flattening for now