Clean up pow tests

[jagerber48]

• Closes Clean up power functions tests #291
• Executed pre-commit run --all-files with no errors
• The change is fully covered by automated unit tests
• Documented in docs/ as appropriate
• Added an entry to the CHANGES file

Summary of goals:

• Changes all power tests to use pytest.mark.parametrize. Hopefully this makes the tests more readable.
• Combines the tests for pow and umath_core.pow in a test_power.py module to get some code re-use and to combine similar tests rather than spread them across modules
• Remove pow test code from helpers.py module.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.71%. Comparing base (19865cc) to head (0b180c9).
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #292      +/-   ##
==========================================
+ Coverage   96.55%   96.71%   +0.15%     
==========================================
  Files          17       18       +1     
  Lines        1946     1886      -60     
==========================================
- Hits         1879     1824      -55     
+ Misses         67       62       -5     

Flag	Coverage Δ
macos-latest-3.10	95.12% <100.00%> (+0.10%)	⬆️
macos-latest-3.11	95.12% <100.00%> (+0.10%)	⬆️
macos-latest-3.12	95.12% <100.00%> (+0.10%)	⬆️
macos-latest-3.8	95.11% <100.00%> (+0.10%)	⬆️
macos-latest-3.9	95.11% <100.00%> (+0.10%)	⬆️
no-numpy	74.60% <100.00%> (-0.53%)	⬇️
ubuntu-latest-3.10	95.12% <100.00%> (+0.10%)	⬆️
ubuntu-latest-3.11	95.12% <100.00%> (+0.10%)	⬆️
ubuntu-latest-3.12	95.12% <100.00%> (+0.10%)	⬆️
ubuntu-latest-3.8	95.11% <100.00%> (+0.10%)	⬆️
ubuntu-latest-3.9	95.11% <100.00%> (+0.10%)	⬆️
windows-latest-3.10	95.12% <100.00%> (+0.10%)	⬆️
windows-latest-3.11	95.12% <100.00%> (+0.10%)	⬆️
windows-latest-3.12	95.12% <100.00%> (+0.10%)	⬆️
windows-latest-3.8	95.11% <100.00%> (+0.10%)	⬆️
windows-latest-3.9	95.11% <100.00%> (+0.10%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

🚀 New features to boost your workflow:

• ❄ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[codspeed-hq]

CodSpeed Performance Report

Merging #292 will not alter performance

_{Comparing jagerber48:pow_tests (0b180c9) with master (19865cc)}

Summary

✅ 5 untouched benchmarks

[newville]

@jagerber48 Thanks - this looks like an improvement in readability to me! I do worry a bit about a test with
integer = ufloat(-3, 0)

It looks like that is only used to test that using that as an exponent yields a stderr of NaN. That's OK.
I would also be OK with deliberately not testing that: the value ufloat(3, 0) makes no sense, so we should feel obligated to promise a sensible result when using it.

+1 on merging.

[newville]

Looks good to me. Thanks!

[jagerber48]

I'll double check that all cases are copied over correctly for now and then merge it in.

I would also be OK with deliberately not testing that: the value ufloat(3, 0) makes no sense, so we should feel obligated to promise a sensible result when using it.

Definitely interested in this idea since it's tests like these that have been giving me a lot of trouble on #262 and caused to me raise #283. We can discuss elsewhere. I'll keep this PR as a simple 1:1 test refactor.