Improve caching

[jm-rivera]

This pull request addresses a bug in the memory cache logic that could return stale or incomplete results when a cached DataFrame did not contain all the columns requested by a subsequent query. The fix ensures that memory cache entries are only used if they include all requested columns, and introduces a projection-specific cache key to store and retrieve column-subset (projected) results separately. This prevents cache pollution and stale data issues for projected reads.

Caching and Query Logic Improvements:

• Updated the memory cache logic in read methods to only return a cached DataFrame if it contains all requested columns, preventing incomplete results from being served. [1] [2]
• Introduced a projection-specific cache key using the new generate_projection_hash function, allowing column-subset (projected) results to be cached and retrieved separately from full-table results. [1] [2]
• Modified bulk cache reads to cache projected DataFrames under their own keys, avoiding pollution of the main filter-based cache and ensuring correct results for future queries with the same projection. [1] [2]

Documentation and Versioning:

• Bumped the package version to 2.4.2 and updated the changelog to document the cache bug fix. [1] [2]

[codecov-commenter]

Codecov Report

❌ Patch coverage is 34.61538% with 17 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (main@3d8795a). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
src/oda_data/api/sources.py	22.72%	14 Missing and 3 partials ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #157   +/-   ##
=======================================
  Coverage        ?   75.64%           
=======================================
  Files           ?       36           
  Lines           ?     2164           
  Branches        ?      198           
=======================================
  Hits            ?     1637           
  Misses          ?      490           
  Partials        ?       37           

Flag	Coverage Δ
unittests	75.64% <34.61%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.