cache getFinishedSync result to eliminate N+1 query pattern

[arreyder]

cache getFinishedSync result in C1File to eliminate N+1 query pattern during pagination
adds sync.Once-guarded cache fields to C1File struct
replaces repeated getFinishedSync calls in listConnectorObjects with cached lookup

production CPU profiling of temporal_sync revealed that getFinishedSync consumed 44.32s (43.11% of CPU) during sync operations. The function was being called on every listConnectorObjects pagination call when reqSyncID is not explicitly set.

ListGrants → listConnectorObjects → getFinishedSync (repeated per page)

I tested this one by reducing the workload page size so I could emulate prod a little better with my tool and I saw good results.

for this case, baton-sdk benchmarks won't show any benefit because they do not work like temporal-sync does when running in prod.

Summary by CodeRabbit

• Chores
  • Improved performance by optimizing how sync data is retrieved during listing operations, reducing unnecessary database queries.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

This pull request adds internal caching to C1File to optimize sync run retrieval and avoid N+1 queries. Three unexported cache fields are introduced alongside a new private method getCachedViewSyncRun that uses sync.Once to cache sync run results. The listConnectorObjects function is updated to use this cached sync instead of directly fetching the latest sync.

Changes

Cohort / File(s)	Change Summary
Cache field initialization pkg/dotc1z/c1file.go	Adds three unexported fields (cachedViewSyncRun, cachedViewSyncOnce, cachedViewSyncErr) to C1File for storing cached sync run results.
Cached sync run method pkg/dotc1z/sync_runs.go	Implements new private method getCachedViewSyncRun(ctx context.Context) that uses sync.Once pattern to cache sync run retrieval, attempting finished full sync first, then falling back to latest unfinished sync.
Query optimization pkg/dotc1z/sql_helpers.go	Updates listConnectorObjects to use getCachedViewSyncRun(ctx) instead of inline sync retrieval, removing the unfinished-sync fallback logic previously present.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• Do grant expansion after compaction. #386: Also modifies C1File's sync-run handling with SetCurrentSync method addition, touching related sync-run logic and the same struct.

Suggested reviewers

• loganintech

Poem

🐰 A cache so fine, a sync run neat,
No more N+1, what a treat!
Once locks the path, so smooth and fleet,
With fallback grace, the loop's complete! ✨

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately reflects the main objective of the PR: adding caching to eliminate N+1 queries on getFinishedSync during pagination.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch crr/cache-view-sync-run

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[ennyjfrick]

you can use sync.OnceValue here instead maybe