test(custom-scheduler): stabilize custom scheduler testing

[mikehardy]

Purpose / Description

After implementing the reviewer fragment thing in the new reviewer with #16455 the custom scheduler test was failing for me on the command line so my customary / frequent ./gradlew jacocoTestReport did not pass

I believe others did not notice it because:

1- it is not executed in CI because of a Flaky annotation
2- I suppose people developing locally either don't run tests or run them from Android Studio? I use the command line

I need that to pass, so I looked into it, and this or something like this will fix it

Fixes

• Fixes regression from feat: start Reviewer port #16455

Approach

1- realize by view hierarchy inspection that the hierarchies are different from the command line and from Android Studio. I have no idea. They are though. So...allow either name for the the layout flip or ease 3 button view resource to match 🤷
2- realize that timing issues are still present (even for me locally) but if we use the otherwise-unused retry rule with a retry of 3, it works every time locally! So I am passing now
3- try to get this working in CI so it doesn't start silently failing again and sending me on a yak shave. Bumping retry to 10 seems to do it? 🙏

Note there was a line-ending issue in initial push that has been peeled off to #16536

How Has This Been Tested?

It's all tests, it is made of tests. But this was a flake, so it has a a higher burden of proof. I'm 9 for 9 running this so I think I actually stabilized it

Learning (optional, can help others)

• New emulator on new ubuntu runner is pretty quick, default view match timeout is 10s and we don't need 30s anymore
• we already had a retry rule which was cool
• view hierarchies can be different between command line and android studio, and I can't even begin to explain that
• I cannot find a way to run just a single test or regex of tests from the androidTest hierarchy via command line 🤷

Checklist

Please, go through these checks before submitting the PR.

• You have a descriptive commit message with a short title (first line, max 50 chars).
• You have commented your code, particularly in hard-to-understand areas
• You have performed a self-review of your own code
• UI changes: include screenshots of all affected screens (in particular showing any new or changed strings)
• UI Changes: You have tested your change using the Google Accessibility Scanner

[mikehardy]

(note, this passes over and over again on my fork (4 in a row now, running 5th) - example - https://github.com/mikehardy/Anki-Android/actions/runs/9341737197/job/25709045233 )

[david-allison]

Rejecting due to whitespace changes in commit 1 on ReviewerTest - the entire file is shown as being recreated

Approve after this is fixed, suggestion is a nitpick for clarity

[mikehardy]

Yeah the spacing change is ridiculous and will definitely be changed if this passes muster at all:

If this works for everyone else I'll fix up the ridiculous "full file line-ending" change thing that happened on my computer. No idea, and I don't want to shave that yak until there's buy in on this PR. View diff with spacing changes ignored

Seems like it might - with nitpick for clarity 👍 - so will clean this up a bit and repush

[david-allison]

Fantastic, cheers!

[mikehardy]

[mikehardy]

Succeeded first time here as well, and the merge queue will run it again and then it'll run again again on push to main, so 3 more runs to give it a final look and hope it is actually non-flaky

If this starts to flake again - for anyone reading this in the future - just readd the Flaky annotations, do not change any of the other work here as it was all necessary (even the retries!) in order for it to work locally even.