MAINT:linalg: Remove redundant det and lu Fortran files

[ilayn]

#18358 finalized replicating the work that these old Fortran files were used for. This PR removes the files themselves.

[j-bowhay]

Opinions on also getting rid of flinalg.py and _flinalg.py too? Last time around there was some concerns about the 2.0.0 deprecation warning. Since then, the only two functions left now have a warning saying they will be removed in this release as they are not part of the linalg namespace which would leave an empty module so seems a little pointless to leave around

In [2]: scipy.linalg.flinalg.__all__
Out[2]: ['get_flinalg_funcs', 'has_column_major_storage']

In [3]: scipy.linalg.flinalg.get_flinalg_funcs
<ipython-input-3-2f187ec6e743>:1: DeprecationWarning: `scipy.linalg.flinalg.get_flinalg_funcs` is deprecated along with the `scipy.linalg.flinalg` namespace. `scipy.linalg.flinalg.get_flinalg_funcs` will be removed in SciPy 1.13.0, and the `scipy.linalg.flinalg` namespace will be removed in SciPy 2.0.0.
  scipy.linalg.flinalg.get_flinalg_funcs
Out[3]: <function scipy.linalg._flinalg_py.get_flinalg_funcs(names, arrays=(), debug=0)>

In [4]: scipy.linalg.flinalg.has_column_major_storage
<ipython-input-4-e9709a88fd27>:1: DeprecationWarning: `scipy.linalg.flinalg.has_column_major_storage` is deprecated along with the `scipy.linalg.flinalg` namespace. `scipy.linalg.flinalg.has_column_major_storage` will be removed in SciPy 1.13.0, and the `scipy.linalg.flinalg` namespace will be removed in SciPy 2.0.0.
  scipy.linalg.flinalg.has_column_major_storage
Out[4]: <function scipy.linalg._flinalg_py.has_column_major_storage(arr)>

[ilayn]

Oh I completely forgot those. Let me check what they do.

[ilayn]

so seems a little pointless to leave around

I agree, removed them now.

[lucascolley]

flinalg also has references in mypy.ini, scipy\_lib\tests\test_public_api.py, scipy\linalg\__init__.py, scipy\linalg\_basic.py, scipy\linalg\basic.py, and scipy\linalg\decomp_lu.py.

edit: happy to make a PR to your branch if you'd prefer

[ilayn]

Oh my if it is not a hassle please go ahead, I don't know how I missed them. Thanks for the drill down though.

[lucascolley]

CI happy (not sure what's up with Cirrus), just need to decide whether

scipy/scipy/linalg/_basic.py

Lines 947 to 955 in e00a594

	# XXX: I found no advantage or disadvantage of using finv.
	# finv, = get_flinalg_funcs(('inv',),(a1,))
	# if finv is not None:
	# a_inv,info = finv(a1,overwrite_a=overwrite_a)
	# if info==0:
	# return a_inv
	# if info>0: raise LinAlgError, "singular matrix"
	# if info<0: raise ValueError('illegal value in %d-th argument of '
	# 'internal inv.getrf|getri'%(-info))

should stay/be reworded/be deleted.

[ilayn]

That comment has no relevance anymore hence I removed it. It was when we tried to use F or C linalg based on the array flag but that's long gone.

[j-bowhay]

Thanks both! Nice to see ~600 lines go

[ilayn]

Thanks @lucascolley for the emergency save, @j-bowhay