fix: normalize totals cache keys for async hits

[betodealmeida]

SUMMARY

Contribution charts were getting cache misses on force_cached=True because the totals query used in post-processing kept its original row_limit (e.g., 1000) when building cache keys on fetch, while the cached entry was produced with row_limit=None during async computation. This mismatch could trigger unnecessary async jobs and redundant totals recalculations.

This change introduces a normalization helper that always sets the totals query row_limit to None before cache-key generation, even for force-cached reads, and reuses that helper inside ensure_totals_available to avoid duplicated logic. The force-cached path now aligns cache keys without executing totals, and a unit test verifies the totals query is normalized without running a new totals query, preserving expected cache hits.

BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF

TESTING INSTRUCTIONS

ADDITIONAL INFORMATION

• Has associated issue:
• Required feature flags:
• Changes UI
• Includes DB Migration (follow approval process in SIP-59)
  • Migration is atomic, supports rollback & is backwards-compatible
  • Confirm DB migration upgrade and downgrade tested
  • Runtime estimates and downtime expectations provided
• Introduces new feature or API
• Removes existing feature or API