Separate release test workflows for PyPI and Conda-forge

[ajjackson]

Closes #346

[ajjackson]

I suppose one benefit of having a test_release that calls both of these workflows would be that it already exists on master and gives a way to run these while working on them...

[github-actions]

Test Results

    20 files  +    6      20 suites  +6   1h 23m 32s ⏱️ + 13m 40s
 1 066 tests ±    0   1 060 ✅ ±    0   6 💤 ± 0  0 ❌ ±0 
12 535 runs  +1 500  12 462 ✅ +1 485  73 💤 +15  0 ❌ ±0 

Results for commit c0aa840. ± Comparison against base commit 62be22d.

♻️ This comment has been updated with latest results.

[ajjackson]

Things seem to be running nicely now https://github.com/pace-neutrons/Euphonic/actions/runs/12401144969

Some concerning Windows failures but I don't think that's the workflow's fault?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.43%. Comparing base (1f3f8c0) to head (c13c818).
Report is 10 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #350   +/-   ##
=======================================
  Coverage   95.43%   95.43%           
=======================================
  Files          29       29           
  Lines        4209     4209           
  Branches      641      641           
=======================================
  Hits         4017     4017           
  Misses        114      114           
  Partials       78       78           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ajjackson]

Hmm, the windows test failures are numerical differences of 0.005%. I'm not thrilled about them, but that's not a "cancel Christmas"-scale problem in my book. We can look at it next year and see if there is some sensitivity to Numpy versions/builds. It will be easier with this test split in place.

[oerc0122]

Looks good. The #349 changes should slot right in, or be unnecessary. We'll find out once this is in.

[github-actions]

PyPI test results

    22 files      22 suites   54m 30s ⏱️
 1 066 tests  1 060 ✅   6 💤 0 ❌
23 320 runs  23 188 ✅ 132 💤 0 ❌

Results for commit c0aa840.

♻️ This comment has been updated with latest results.

[oerc0122]

Just some minor personal legibility things and symmetry. None of which are probably important.

[github-actions]

Conda-forge test results

    12 files      12 suites   31m 35s ⏱️
 1 029 tests  1 019 ✅  5 💤  5 ❌
12 276 runs  12 201 ✅ 60 💤 15 ❌

For more details on these failures, see this check.

Results for commit c0aa840.

♻️ This comment has been updated with latest results.

[ajjackson]

The checks now look correct: PyPI pass and Conda-forge fail appear separately 😄