CLN: Enforce deprecation of using alias for builtin/NumPy funcs

[rhshadrach]

• closes #xxxx (Replace xxxx with the GitHub issue number)
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

Ref: #53974

[rhshadrach]

A bit of a complication here - I missed one of the paths in apply where we use the builtin/NumPy -> cython table.

pandas/pandas/core/apply.py

Lines 1439 to 1442 in 54d2033

	if callable(func):
	f = com.get_cython_func(func)
	if f and not self.args and not self.kwargs:
	return obj.apply(func, by_row=False)

Sadly, it's probably one of the more common cases where users will see a difference. This is because apply will first try to operate row-by-row (unless by_row=False), and e.g. np.sum works on rows.

I'm thinking we enforce the deprecation as-is (we already were not very consistent with where we were making these replacements), and deprecate the last remaining case as part of 3.x.

cc @jorisvandenbossche @mroeschke

[mroeschke]

Changes generally look good. npdev failure looks real?

I'm thinking we enforce the deprecation as-is (we already were not very consistent with where we were making these replacements), and deprecate the last remaining case as part of 3.x.

Yeah fine to do a follow up deprecation in 3.x

[jorisvandenbossche]

Yes, sounds good to me too!

[rhshadrach]

@mroeschke - green now.

[mroeschke]

Thanks @rhshadrach