fix rebasing cycle heads when not reaching a fixpoint

[lcnr]

fixes rust-lang/trait-system-refactor-initiative#232

annoyingly subtle, imagine the following proof tree

• A (no cycle head usages, final result Y)
  • ignored B (depends on A with provisional result X)
    • A (cycle, provisional result X)
• B (using the cache entry here incorrectly assumes A has final result X)

r? @BoxyUwU

[lcnr]

moved earlier for potential perf gains. no need to do anything more involved if we need to pop anyways

[lcnr]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

[rustbot]

This PR was rebased onto a different master commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[lcnr]

@bors cancel

@bors try

[BoxyUwU]

r=me if ci and perf good

[rust-bors]

☀️ Try build successful (CI)
Build commit: a7a478b (a7a478bb1c36795d5c8eb82ffec98b5b4200479d, parent: 5b9007bfc358817cf066ee27c6b440440727d3a7)

[rust-timer]

Queued a7a478b with parent 5b9007b, future comparison URL.
There are currently 0 preceding artifacts in the queue.
It will probably take at least ~1.1 hours until the benchmark run finishes.