Implement `yarn lint --fix` differently, and rework scripts

[mstange]

WIP

[codecov[bot]]

Codecov Report

:white_check_mark: All modified and coverable lines are covered by tests. :white_check_mark: Project coverage is 85.73%. Comparing base (c405844) to head (8dd8df7). :warning: Report is 89 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #5595   +/-   ##
=======================================
  Coverage   85.73%   85.73%           
=======================================
  Files         309      309           
  Lines       30390    30390           
  Branches     8365     8365           
=======================================
  Hits        26055    26055           
  Misses       3914     3914           
  Partials      421      421           

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

:rocket: New features to boost your workflow:

• :snowflake: Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• :package: JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[canova]

Hmm it looks like a lot of complexity for this. TBH I'm not so sure if it's worth it.

[mstange]

I'm planning to write up a bunch of before/after examples in the summary, then we can discuss if it's worth it :)

It looks complex because:

• one script gets split into three (whose total line count is a bit bigger than the original script's line count, but not by much)
• a bunch of build-internal commands are added - I'll investigate this again, I think I was solving an issue that was orthogonal to the original intent

Specifically I was annoyed that you got a suggestion to run yarn test -u even though you're already running yarn test -u, if there's a real test failure.