feat: optimize Horreum data loading with concurrency

[deekshith-24]

Summary

Optimizes Horreum data loading by implementing HTTP connection pooling and concurrent processing.

Performance improvement: ~19x faster

Changes

1. HTTP Session Pooling - Reuses TCP connections instead of creating new ones for each request
2. Concurrent Trashed Run Checking - Checks 5 runs in parallel instead of sequentially
3. Concurrent Alerting Variable Checking - Checks all alerting variables in parallel

Technical Details

• Added requests.Session() for connection pooling
• Used ThreadPoolExecutor for concurrent processing (5 workers for trashed checks, 10 for alerting)
• Replaced all requests.get/post/put/delete with session methods
• Early termination when duplicate/change is found

Testing

• ✅ Validated with 75 successful uploads
• ✅ Measured average upload time: 10.7 seconds (down from ~120+ seconds)
• ✅ Result determination: 5.9 seconds (down from ~115 seconds)

Co-Authored-By: Claude Sonnet 4.5 noreply@anthropic.com