Fix config test isolation on release runners

[patchstackdave]

Summary

Fix the config test that passes in pull-request CI but fails on main-branch and release runners.

The fixture inherited GitHub hosting signals. A committed sandbox label intentionally does not override a production platform build, so the release runner correctly resolved production while the test expected sandbox.

• Give config fixtures an isolated environment and explicit hosting signals.
• Add 18 regression cases covering main-branch dispatch/push, tag releases, feature branches, pull requests, and pull-request-target events.
• Verify committed local/sandbox labels, process overrides, provenance, and unchanged config files.
• No runtime behavior changes, skipped assertions, or weakened publish gates.

Validation

• Reproduced the original failure with GitHub main-branch workflow-dispatch variables.
• Config/environment tests passed under those variables; config tests also passed under tag-release variables.
• Full suite under the main-branch release environment: 4,276 passed, 7 skipped.
• Typecheck (10 templates), build, mandatory security canary (7 tests), and npm pack dry-run passed.
• Public-boundary review completed; fixtures use synthetic values.

Release follow-up

Existing release tags are unchanged. Publish from a new tag containing this fix after merge; rerunning the original v0.5.29 tag retains its failing test. No package has been published by this change.

[coderbuds]

Tests now fully isolate environment variables with cleaner setup.

🎯 Quality: 94% Elite · 📦 Size: Small

📈 This month: Your 205th PR — above team average · Averaging Excellent

_{See how your team is trending →}

[patchstackdave]

/review