refactor: add copy helpers

[agoose77]

No description provided.

[codecov]

Codecov Report

Merging #1697 (5f08a37) into main (e692946) will increase coverage by 0.43%.
The diff coverage is 87.18%.

❗ Current head 5f08a37 differs from pull request most recent head 355995d. Consider uploading reports for the commit 355995d to get more accurate results

Additional details and impacted files

Impacted Files	Coverage Δ
src/awkward/_v2/_connect/jax/__init__.py	90.47% <ø> (+1.58%)	⬆️
src/awkward/_v2/config.py	0.00% <0.00%> (ø)
src/awkward/_v2/operations/ak_broadcast_arrays.py	100.00% <ø> (+4.34%)	⬆️
src/awkward/_v2/operations/ak_transform.py	65.51% <ø> (+56.89%)	⬆️
src/awkward/_v2/types/uniontype.py	84.61% <ø> (ø)
src/awkward/_v2/behaviors/categorical.py	73.13% <60.00%> (-8.94%)	⬇️
src/awkward/_v2/forms/form.py	82.95% <68.51%> (-7.80%)	⬇️
src/awkward/_v2/highlevel.py	72.09% <72.72%> (+0.25%)	⬆️
src/awkward/_v2/forms/emptyform.py	76.47% <75.00%> (+1.08%)	⬆️
src/awkward/_v2/forms/unionform.py	73.75% <75.00%> (+0.57%)	⬆️
... and 51 more

[agoose77]

Given that we just don't allow cycles in parameters, this is not required.

[jpivarski]

Okay, so that was the motivation for this!

This was looking to me like an abstraction of __deepcopy__ from a computational implementation, "deepcopy does such-and-such an algorithm," to a declarative specification: "these attributes are deep and those are shallow, a single algorithm for all layouts uses this fact." I didn't think this abstraction was necessary if it was only for __deepcopy__, but if there was some other reason for it, then maybe it would help to have this information around.

But if your goal was just for __deepcopy__ and you don't have any other operations in mind, then I agree that this is probably overkill.