fix(ci): refetch PR before labeling new contributors

[TomCC7]

Contribution path

Small, safe change that does not need a tracking issue.

Problem

The first-time-contributor job trusted author_association from the pull_request_target.opened payload. For PR #3226, that snapshot reported NONE, while a fresh pull request API response reported FIRST_TIME_CONTRIBUTOR. The job returned early and skipped the label.

Solution

Refetch the pull request with pulls.get before evaluating author_association. The job still runs only for opened, retains the label as a snapshot, executes no PR code, and keeps its existing narrow permissions.

How to Test

uv run python -c 'import pathlib, yaml; assert "first-time-contributor" in yaml.safe_load(pathlib.Path(".github/workflows/pr-labels.yml").read_text())["jobs"]'

Also compiled the embedded github-script as an async function and exercised FIRST_TIMER, FIRST_TIME_CONTRIBUTOR, and NONE with a mocked stale opened-event payload. git diff --check passes. PR #3226 was backfilled with the label.

AI assistance

Codex with GPT-5 diagnosed the CI failure, implemented the workflow change, and ran the validation harness. The author reviewed the requested fix and PR creation.

Checklist

• I have read and approved the CLA.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ All tests successful. No failed tests found.

@@            Coverage Diff             @@
##             main    #3229      +/-   ##
==========================================
- Coverage   74.90%   74.89%   -0.01%     
==========================================
  Files        1114     1114              
  Lines      106086   106086              
  Branches     9665     9665              
==========================================
- Hits        79459    79456       -3     
- Misses      23817    23820       +3     
  Partials     2810     2810              

Flag	Coverage Δ
OS-ubuntu-24.04-arm	68.50% <ø> (-0.01%)	⬇️
OS-ubuntu-latest	70.51% <ø> (-0.01%)	⬇️
Py-3.10	70.51% <ø> (-0.01%)	⬇️
Py-3.11	70.51% <ø> (-0.01%)	⬇️
Py-3.12	70.51% <ø> (-0.01%)	⬇️
Py-3.13	70.51% <ø> (-0.01%)	⬇️
Py-3.14	70.51% <ø> (-0.01%)	⬇️
Py-3.14t	70.51% <ø> (-0.01%)	⬇️
SelfHosted-Large	29.33% <ø> (+<0.01%)	⬆️
SelfHosted-Linux	35.93% <ø> (ø)
SelfHosted-macOS	35.02% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.
see 3 files with indirect coverage changes

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[Dreamsorcerer]

Not obvious to me this is the correct cause of that failure, but difficult to test without merging.

[TomCC7]

Not obvious to me this is the correct cause of that failure, but difficult to test without merging.

yeah ci is so annoying to test orz