Browsertime geckoview tests are no longer running because of this patch: https://phabricator.services.mozilla.com/D74306

Looks like my careful taskgraph-diffing missed this between the land backout and reland, (afaict) -- https://phabricator.services.mozilla.com/D74306?vs=280735&id=282615#toc specifically was the rebase that didn't catch this regression.

Perf team uses https://github.com/gmierz/moz-current-tests/blob/master/generate_test_report.py to vet the list of test changes before/after, while I use taskgraph-gen from https://hg.mozilla.org/build/braindump/file/tip/taskcluster/taskgraph-diff/

In this particular case I also have a performance cron param file locally to diff against to (I hoped) avoid this very sort of issue, and I was using a combination of gron and cat ... | jq 'keys' to help vet the before/after state --> which for this patch set wasn't the easiest since I had to compare label name changes too.

Overall intent I had going in is that I wouldn't be removing any job that ran on the 'pgo' jobs, just moving them to the new naming 'shippable'

Yea I think this was from one of those cycles, it's possible something was missed since we were also landing the browsertime geckoview very close to the same time as that one. Here's a diff of the changes that browsertime underwent in the last week (since bug 1623355 landed): https://www.diffchecker.com/boZzmxAc

This patch fixes the run-on-projects settings for browsertime mobile tests which were regressed from a bad merge (or multiple large patches landing very close to each other).

Set release status flags based on info from the regressing bug 1623355

https://hg.mozilla.org/mozilla-central/rev/b6b7c436d0c9