Adds a structured set of 11 query scenarios covering the main catalog
database access patterns: paginated listings, counts, facets, entity
lookups, full-text search, ancestry traversal, stitching reference
counts, and orphan detection.
Each scenario documents the SQL, what a healthy plan looks like, and
what anti-patterns to watch for. The baseline records execution times
and plan shapes from the staging database (545K entities, 13.8M search
rows).
This is intended to be run by humans or AI agents before and after
database-affecting changes to detect performance regressions. It lives
alongside the existing performance tests.
Signed-off-by: Fredrik Adelöw <freben@gmail.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Fredrik Adelöw <freben@gmail.com>
Fredrik Adelöw
d2662df5de