fix: Reset ResourceSyncFailing condition to False on successful reconcile

[wikoion]

Sinker sets ResourceSyncFailing=True when a reconcile fails but never reliably clears it on success, so a single transient error can leave the condition latched forever. We've had alerts firing in starfleet on syncs that recovered seconds after the failure.

• Successful reconciles now write an explicit ResourceSyncFailing=False condition instead of nulling the status. The old success path compared the computed status against the cached object, which skips the clearing patch when the cache is stale.
• lastTransitionTime is only carried over while the condition value is unchanged, so a True<->False flip records a real transition.
• Deleted resources keep their existing status so we don't patch objects whose finalizer has just been removed.
• Reworked the transition time tests, the old assertion passed on any negative diff so the existing-condition case wasn't asserting anything.

[Copilot]

🟡 Changes recommended

The core success and deletion status-selection behavior lacks regression coverage.

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Ensures recovered ResourceSync reconciliations clear stale failure conditions.

Changes:

• Writes explicit success and failure conditions.
• Preserves transition times only when status is unchanged.
• Expands transition-time tests.

File summaries

File	Description
src/controller.rs	Updates condition handling and tests.

Review details

• Files reviewed: 1/1 changed files
• Comments generated: 1
• Review effort level: Balanced

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.