ref(releases): Bulk read and update on the happy path

[cmanallen]

Presently we are reading then updating each release individually. This is inefficient. We can read and update multiple releases at a time. As of the most recent run we read and updated 100x the number of rows in the happy path as we did the unhappy path. By implementing the bulk logic in the happy path first we can minimize our change surface and receive 99% of the reward. Note: tasks are chunked by organization so we can't move everything into a single query. But for the scope of an individual task (org) we are reading and updating with two queries.

Moving to this bulk refactor allowed us to further flatten the function definitions. Making them more tractable when trying to read and understand the code. This new behavior has unit and integration testing in addition to the unchanged functional test coverage. In a follow up PR I'll clean up the unhappy path.

Current total system time is ~15 minutes. We'll measure success by stable operation and improvements to total system time.

[codecov]

Codecov Report

Attention: Patch coverage is 93.60000% with 8 lines in your changes missing coverage. Please review.

✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
src/sentry/release_health/tasks.py	86.66%	8 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master   #92508      +/-   ##
==========================================
+ Coverage   87.78%   87.92%   +0.14%     
==========================================
  Files       10258    10250       -8     
  Lines      588215   587705     -510     
  Branches    22859    22727     -132     
==========================================
+ Hits       516347   516742     +395     
+ Misses      71421    70541     -880     
+ Partials      447      422      -25     

[aliu39]

Changes look good!