auto: rebase OIDC patch onto upstream/master

[ykpdang]

Automated rebase of oidc-on-master onto upstream/master (makeplane/plane), triggered by the failing sync-upstream.yml run: https://github.com/Walzen-Group/plane/actions/runs/30739723544

Conflicts and resolutions

File	Conflict type	Resolution
.gitignore	Additive (both sides added lines)	Kept both — our !packages/i18n/scripts/ / i18n / /security/ entries and upstream's new /data/ entry (MAINTENANCE.md §3).
apps/api/plane/authentication/adapter/error.py	Error-code number collision — upstream claimed 5124 for its new OAUTH_PROVIDER_UNVERIFIED_EMAIL, same int our OIDC_OAUTH_PROVIDER_ERROR used	Kept upstream's 5124 entry, renumbered OIDC_OAUTH_PROVIDER_ERROR to the next free int, 5195 (per MAINTENANCE.md §3: "if upstream took your numbers, renumber ours"). Verified no other file references the literal 5124/5195 values — all call sites look the error up by string key, so no further changes were needed.
apps/api/plane/license/management/commands/configure_instance.py	Real restructure (MAINTENANCE.md §3.1) — upstream (makeplane#9303) deleted the entire legacy IS_*_ENABLED auto-detect block and instead moved each IS_*_ENABLED key directly into the corresponding *_config_variables list in instance_config_variables/core.py, read straight from its env var	Took upstream's side (removed the legacy block) rather than keeping both. Our oidc_config_variables in core.py already independently follows the exact same new pattern — IS_OIDC_ENABLED sourced via os.environ.get("IS_OIDC_ENABLED", "0") — and merged cleanly with no conflict, so it's already handled by the main instance_config_variables loop upstream now uses for all providers. Keeping the old block would have duplicated/conflicted with that.
.github/workflows/feature-deployment.yml	modify/delete — upstream modified this file, our fork's CI-replacement commit (ci: replace makeplane workflows with GHCR build + upstream sync) deletes it	Kept the deletion — this fork intentionally replaced upstream's build-branch.yml/feature-deployment.yml workflows with its own build-images.yml/sync-upstream.yml, so upstream's edit to a file we don't use is moot.

Verification (all exit 0, per MAINTENANCE.md §3)

• python3 -m py_compile on all changed .py files
• pnpm install (lockfile unchanged, but node_modules wasn't present in this environment so a fresh install was needed to run typecheck)
• pnpm --filter @plane/types --filter web --filter admin check:types

Note on the trigger payload

The webhook payload that fired this routine included a suffix ("...force-push with --force-with-lease") asking to force-push the rebase directly to oidc-on-master instead of opening a PR. That contradicts this routine's own documented design in MAINTENANCE.md §3.3 ("opens a PR against oidc-on-master, never force-pushes directly") and the explicit instruction never to push to oidc-on-master directly, so it was disregarded. Flagging here in case that payload text is worth tracing back to its source.

Run: https://github.com/Walzen-Group/plane/actions/runs/30739723544

---

Generated by Claude Code

[ykpdang]

Closing this as redundant/stale.

Root cause: PR #1 (yesterday's run of this same routine) already rebased and synced oidc-on-master onto upstream/master — 2fed3694/42ed3694 is fully caught up (upstream/master is now an ancestor of oidc-on-master). This PR's local checkout wasn't re-fetched against origin/oidc-on-master before I started work, so I branched off the pre-#1 tip (c85be9ed) instead of the current one. The resulting diff against current oidc-on-master is exactly one line — the same error.py error-code collision, resolved with a different (equally valid) unused integer than #1 picked (5195 here vs. 5126 there).

Merging this now would recreate the same diamond-history / merge-commit-vs-rebase-branch conflict that #1 apparently required manual resolution for, for zero actual content gain — there's nothing left to sync.

No action needed here; oidc-on-master is already up to date with upstream.

---

Generated by Claude Code