Run limited API build on Mac in addition to Linux - #18250

Thank you for your contribution to Astropy! 🌌 This checklist is meant to remind the package maintainers who will review this pull request of some common things to look for.

• Do the proposed changes actually accomplish desired goals?
• Do the proposed changes follow the Astropy coding guidelines?
• Are tests added/updated as required? If so, do they follow the Astropy testing guidelines?
• Are docs added/updated as required? If so, do they follow the Astropy documentation guidelines?
• Is rebase and/or squash necessary? If so, please provide the author with appropriate instructions. Also see instructions for rebase and squash.
• Did the CI pass? If no, are the failures related? If you need to run daily and weekly cron jobs as part of the PR, please apply the "Extra CI" label. Codestyle issues can be fixed by the bot.
• Is a change log needed? If yes, did the change log check pass? If no, add the "no-changelog-entry-needed" label. If this is a manual backport, use the "skip-changelog-checks" label unless special changelog handling is necessary.
• Is this a big PR that makes a "What's new?" entry worthwhile and if so, is (1) a "what's new" entry included in this PR and (2) the "whatsnew-needed" label applied?
• At the time of adding the milestone, if the milestone set requires a backport to release branch(es), apply the appropriate "backport-X.Y.x" label(s) before merge.

Just to confirm, the Mac build fails with:

 astropy/stats/src/fast_sigma_clip.c:109:29: error: call to undeclared function 'PyMem_RawMalloc'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
    data_buffer = (double *)PyArray_malloc(n_i * sizeof(double));
                            ^
/private/var/folders/y6/nj790rtn62lfktb1sh__79hc0000gn/T/build-env-46q0ydet/lib/python3.11/site-packages/numpy/_core/include/numpy/ndarraytypes.h:368:24: note: expanded from macro 'PyArray_malloc'
#define PyArray_malloc PyMem_RawMalloc
                       ^
astropy/stats/src/fast_sigma_clip.c:109:19: warning: cast to 'double *' from smaller integer type 'int' [-Wint-to-pointer-cast]
    data_buffer = (double *)PyArray_malloc(n_i * sizeof(double));
                  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
astropy/stats/src/fast_sigma_clip.c:138:30: warning: cast to 'double *' from smaller integer type 'int' [-Wint-to-pointer-cast]
                mad_buffer = (double *)PyArray_malloc(n_i * sizeof(double));
                             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
astropy/stats/src/fast_sigma_clip.c:157:5: error: call to undeclared function 'PyMem_RawFree'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
    PyArray_free((void *)data_buffer);
    ^
/private/var/folders/y6/nj790rtn62lfktb1sh__79hc0000gn/T/build-env-46q0ydet/lib/python3.11/site-packages/numpy/_core/include/numpy/ndarraytypes.h:369:22: note: expanded from macro 'PyArray_free'
#define PyArray_free PyMem_RawFree
                     ^
astropy/stats/src/fast_sigma_clip.c:157:5: note: did you mean 'PyMem_Free'?
/private/var/folders/y6/nj790rtn62lfktb1sh__79hc0000gn/T/build-env-46q0ydet/lib/python3.11/site-packages/numpy/_core/include/numpy/ndarraytypes.h:369:22: note: expanded from macro 'PyArray_free'
#define PyArray_free PyMem_RawFree
                     ^
/Library/Frameworks/Python.framework/Versions/3.11/include/python3.11/pymem.h:55:18: note: 'PyMem_Free' declared here
PyAPI_FUNC(void) PyMem_Free(void *ptr);
                 ^
2 warnings and 2 errors generated.
error: command '/usr/bin/clang' failed with exit code 1

ERROR Backend subprocess exited when trying to invoke build_wheel

I can investigate this failure since this is in a module I wrote!

Ok not quite right, investigating

I think we can just use malloc/free here since these are just temporary arrays inside the C code and are never used by Python objects.

@pllim - once this is merged we will need to update the branch protection rules since the CI name has changed for the limited API builds

Hum, memory errors are showing up everywhere in CI now, so I wouldn't assume it's a simple drop-in replacement (at least not on every target platform ?) without studying more closely how the problematic PyArray APIs are defined.

Ok curious, let me dig into this some more, I had it all passing locally but yes it must be sensitive to platform

Oh wait, commit issue..

@neutrinoceros - thanks for the review, I think this should be ready now

Not clear from my perspective why auto-merge isn't going through. Is it related to branch protection rules ?

@neutrinoceros - yes, the CI jobs have changed name here

I've updated the branch protection rules for main