Exclude masked weights from average combinations

[nomad3]

Summary

• exclude weights belonging to masked, clipped, or non-finite samples from the weighted-average denominator
• preserve the existing weighted numerator, custom callback behavior, weighted sums, and clean-stack allocation path
• add public and direct regression coverage for frame and pixel weights, clipping, fully masked output, custom callbacks, and weighted sums

Fixes #893

Validation

• python -m pytest ccdproc -q — 337 passed, 29 skipped
• python -m pytest ccdproc/tests/test_combiner.py -q — 80 passed
• NumPy weight-focused tests — 12 passed
• Dask weight-focused tests without Bottleneck — 12 passed
• JAX weight-focused tests with 64-bit enabled — 12 passed
• ruff check ccdproc/combiner.py ccdproc/tests/test_combiner.py
• black --check ccdproc/combiner.py ccdproc/tests/test_combiner.py
• git diff --check

The array-api-strict backend was not counted as patch validation because current main fails earlier in existing Combiner construction and public combine() code. PR #925 changes an adjacent dtype-conversion line; this PR deliberately leaves that conversion untouched.

AI assistance disclosure

I used OpenAI Codex to assist with repository inspection, test design, implementation, and review. I reviewed the resulting changes and take responsibility for their correctness and maintenance.

Checklist

• For new contributors: Did you add yourself to the AUTHORS.rst file?
• For documentation changes: Does your commit message include a [skip ci]? (Not applicable.)
• Did you add an entry to the CHANGES.rst file?
• Did you add a regression test?
• Does the commit message include a Fixes #issue_number?
• Does this PR add, rename, move or remove any existing functions or parameters? (No.)

[Copilot]

Pull request overview

This PR fixes weighted-average combination behavior so that masked/clipped/non-finite samples don’t incorrectly contribute their weights to the weighted-average denominator, aligning combine(..., method="average", weights=...) with expectations for masked/clipped data.

Changes:

• Adjust weighted-average computation to zero out weights corresponding to masked/clipped (NaN-substituted) samples before forming the denominator.
• Add regression tests covering frame weights, per-pixel weights, clipping, fully-masked output, custom callbacks, and weighted sums.
• Update release notes and contributor list.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
ccdproc/combiner.py	Excludes weights at NaN-substituted (masked/clipped) positions from the weighted-average denominator.
ccdproc/tests/test_combiner.py	Adds direct regression tests for weighted-average behavior with masking/clipping and related edge cases.
CHANGES.rst	Documents the bug fix in the unreleased changelog section.
AUTHORS.rst	Adds a new contributor entry.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.63%. Comparing base (313d744) to head (933610c).
⚠️ Report is 10 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #952      +/-   ##
==========================================
+ Coverage   96.56%   96.63%   +0.07%     
==========================================
  Files           8        8              
  Lines        1571     1574       +3     
==========================================
+ Hits         1517     1521       +4     
+ Misses         54       53       -1     

☔ View full report in Codecov by Harness.
📢 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.

[mwcraig]

Thanks @nomad3 -- this looks great, just one minor change so that we do not create an additional array API issue down the road.

[mwcraig]

Thanks for the PR< and apologies for the delayed merge!