Release 4.1.2

[turegjorup]

Summary

Roll up the accumulated [Unreleased] changes since 4.1.1 into a tagged release.

PATCH-level per strict SemVer reading: every change is a bug fix, documentation accuracy correction, or CI/tooling improvement. No new public API; no removed, renamed, or re-typed public API. The runtime behaviour change ($previous chained at five catch sites) is strictly additive — consumers that traverse getPrevious() now see the originating Guzzle/firebase/PSR exception where they previously saw null.

What's in 4.1.2

Fixed

• Consistent $previous chaining in OpenIdConfigurationProvider catch blocks (validateIdToken, getJwtVerificationKeys, fetchJsonResource ×2, getConfiguration).
• Broken CI build-status badge in README.md (was pointing at non-existent pr.yaml and stale "Test & Code Style Review" workflow name).

Changed (documentation accuracy)

• Tightened @throws PHPDoc on validateIdToken, getIdToken, getBaseAuthorizationUrl to enumerate the actual transitive exceptions. Removed the inaccurate @throws ClientExceptionInterface declaration on getIdToken (the body's catch-all wraps it as CodeException with the original chained — that exception never actually escaped this method).

Documentation

• Documented HTTP timeout / proxy / verify configuration via constructor $options in README.md. Capability already provided by league/oauth2-client; no code change.

Tooling / CI

• Bumped actions/checkout v5 → v6 across all workflows.
• Added ci profile to docker-compose matrix services so they don't start during local dev.
• Fixed test:coverage to run via docker-compose with XDEBUG_MODE=coverage.
• Fixed test:run to remove stale composer.lock before composer update.
• Fixed test:matrix:reset to use --profile ci.
• Removed unused .markdownlint.json.

Backward compatibility

Every existing consumer that compiled and ran against 4.1.1 continues to work against 4.1.2 unchanged. Consumers reading $e->getPrevious() on exceptions thrown from these five catch sites will now see a non-null cause where they previously saw null — strict improvement.

Test plan

• task lint:markdown — CHANGELOG and README clean
• task test:coverage — 100% coverage maintained (24/24 methods, 148/148 lines)
• task analyze:php — PHPStan max, no errors
• task lint:php — clean
• CI matrix on PR

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (6aa46d4) to head (2df37c3).

Additional details and impacted files

@@             Coverage Diff             @@
##                main       #39   +/-   ##
===========================================
  Coverage     100.00%   100.00%           
  Complexity        62        62           
===========================================
  Files              1         1           
  Lines            172       172           
===========================================
  Hits             172       172           

Flag	Coverage Δ
unittests	100.00% <100.00%> (ø)

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.