Take trevato/forge-upstream's main (979e19e) on today's main, with every test file in exactly one gate list #1473

Replaces #1303 (a person's ask). Both tries went red on the gate's list check. As pull #1376 found, the cause was main itself, which had four test files listed nowhere. 0.7.75 lists or excludes them, so the take can now be measured on its own.

Done looks like.

• [ ] refs/following/main (979e19e, or the newer one if it moved) is merged into a branch cut from today's main.
• [ ] Where the two cannot both be kept (src/checks/service.ts, test/check-tiers.test.ts and any others), this repository's direction wins. The pull lists each conflict and the side kept.
• [ ] Each test file the upstream adds is either in GATE or in EXCLUDED with a reason and an issue number, so test/check-list.test.ts passes.
• [ ] Before the pull, bunx tsc --noEmit, bun test test/check-list.test.ts and the tests for every file the merge touched pass in the sandbox, and the pull names them.
• [ ] Afterwards this repository and the upstream agree up to 979e19e: the following page no longer counts those 14 changes.

Files. whatever the merge touches; scripts/gate.ts. Fits: mason.

Parked by the crew: the run ended without a pull: declined: I didn't push anything because there is nothing left to merge. Today's main (5b33981f, release 0.7.75) already contains the upstream's work, so this pull would have no changes.

Parked by the crew: the run ended without a pull: declined: I've pushed nothing again because there is still nothing to merge. The only new thing on the issue is the note that parked it; the code hasn't moved since the last run.

Parked twice since a person last spoke here, so it is the planner's now. Every reason:

1. the run ended without a pull: declined: I didn't push anything because there is nothing left to merge. Today's main (5b33981f, release 0.7.75) already contains the upstream's work, so this pull would have no changes
2. the run ended without a pull: declined: I've pushed nothing again because there is still nothing to merge. The only new thing on the issue is the note that parked it; the code hasn't moved since the last run

The planner splits it, replaces it with a fresh issue, or defers it; a person can also say what to try. Nothing starts here until then, and the board lists it.

replaced by #1482. Main already contains forge-upstream's 979e19e (git merge-base --is-ancestor 979e19e main holds), so there is nothing left to merge, and both runs rightly declined. The defect is that the converge issue never closes itself once the fork holds the parent's tip. #1482 fixes that, so this kind of issue stops reaching the crew after it is done.