Key Findings:
With the previous bottleneck high page faults resolved in jemalloc (#80245), we have identified a 76% hotspot in performance cycles native_queued_spin_lock_slowpath from QueryConditionCache::write in the latest build on the 2 x 240 vCPUs system. And we have discovered the absence of checks for mark_ranges and has_final_mark, causing all threads to attempt locking entry->mutex unnecessarily.

Solution:
Introduced a new method to QueryConditionCache to determine if cached query conditions need updating based on changes to underlying data or configuration.
This method avoids unnecessary cache rebuilds, ensuring data consistency and reducing lock contention, thereby improving concurrency in multi-threaded query environments.

Performance improvements:

• Reduces the native_queued_spin_lock_slowpath hotspot of entry->mutex from 76% to 1% with Clickbench Q10 on a 2 x 240 vCPUs system
• Increases QPS for Q10 and Q11 by 85% and 89% respectively
• Improves overall performance across 43 queries with a geometric mean gain of 8.1%

Changelog category (leave one):

• Performance Improvement

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

Avoid unnecessary update and reduce lock contention in QueryConditionCache

Documentation entry for user-facing changes

• Documentation is written (mandatory for new features)