chore: Bypass Firefox for Playwright test due to flakiness

[holloway]

These tests don't test anything Firefox-specific (it's conventional JavaScript mostly dependant on a 3rd-party library) so we've decided to bypass these tests in Firefox and just rely on Chrome.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.89%. Comparing base (c7f6bde) to head (59bc498).
Report is 96 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8050      +/-   ##
==========================================
+ Coverage   88.78%   88.89%   +0.10%     
==========================================
  Files         296      304       +8     
  Lines       41320    41302      -18     
==========================================
+ Hits        36687    36715      +28     
+ Misses       4633     4587      -46     

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

[holloway]

fyi @rjsparks

[jennifer-richards]

Is there a way to simply remove Firefox from the list being tested, or is skipping it the only way to do that?

I think part of our motivation for testing against firefox was to ensure we're not using corners that are browser-dependent. Not for this PR, but is there some alternative way to test that?

[rjsparks]

I think we had also hoped that testing against Firefox and Chrome would also give us an early warning if either browser changed behavior in a way it shouldn't that we would need to adjust for (this has happened to us at several meetings).

I agree that right now the firefox playwright tests are not helping us because of the random inexplicable failures. I'm ok with muting that noise for the moment, but I think we should put pressure on making sure we're not surprised by activity at firefox.

[NGPixel]

One reason to keep the Firefox tests is to make sure we don't start using a CSS or JS feature that's not yet supported in Firefox. Firefox tends to be behind times when it comes to implementing new standards.

[holloway]

My understanding is that only the status.spec.js test is flakey in Firefox, so I've tried to narrowly disable only that.

[rjsparks]

merged, but I have to wonder if it was telling us there's a problem with the status notifications on Firefox?