feat(backend): optimize sql queries in Audit Logger

[emmanuelStack654]

closes #904

This pull request optimizes the SQL query performance within the Audit Logger service (auditService.js) and helper (audit.js). Previously, the paginated log retrieval relied on a COUNT(*) OVER() window function to retrieve the total count and log data in a single round-trip. While this avoided a round-trip, it forced PostgreSQL to scan and sort all matching logs for a merchant before applying the limit and offset. In production databases with large volumes of log records, this results in significant performance degradation.

To resolve this issue, the retrieval logic has been restructured to query the database count and log records in parallel using Promise.all. The count query performs a quick index-only scan on the composite (merchant_id, timestamp) index, and the logs query utilizes a backward index scan that terminates immediately after retrieving the specified page limit. Furthermore, all query interactions have been migrated to the optimized database pooler (optimizedQuery and optimizedWrite), integrating query result caching, write invalidation, database-level rate limiting, and parameter signature verification.

Unit tests in auditService.test.js and audit.test.js have been updated to align with the parallelized query execution model. The Vitest database mock was extended to mock both direct pool queries and retried queries, and assertions were updated to verify the separate count and data fetch patterns.

[vercel]

@emmanuelStack654 is attempting to deploy a commit to the Emmanuel's projects Team on Vercel.

A member of the Team first needs to authorize it.

[drips-wave]

@emmanuelStack654 Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits