Delete the 18 consumed gunbc#11056 BuildPath* admission rows

[gunbai-bot]

Summary

• Deletes the 18 TargetChanged admission rows authored by gunbc#11056 for the BuildPath* relocation (ancestry → generation).
• Those rows are consumed by Native ancestry: seed is never the miss path #11056 merging: the base then binds each spelling to v2.compiler.self_host.generation, so the deltas stop being producible. Fierce-lark: open the deletion follow-up at the same time; it lands immediately after Native ancestry: seed is never the miss path #11056 (head of the remaining order).
• Diff is this deletion only (base is session/proud-carp-305). After Native ancestry: seed is never the miss path #11056 merges, retarget to main.
• Does not touch Bind the Filesystem and Clock service reads to the service that answers them #11156's two rows (Delete the two consumed gunbc#11156 admission rows #11240).

Test plan

• After Native ancestry: seed is never the miss path #11056 is on main, required floor namespace-wave-admission reports these 18 as consumed, then green after this deletion
• Join each deleted row against main on (module, in_declaration, spelling, target)

Made with Cursor

[gunbai-bot]

review 65329: the missing dissolution paragraph is the right shape of record, but it cannot be written on this head.

The rows' own trigger is #11056 merges. #11056 is still OPEN (efc1fe04a7a, not on origin/main; git log --oneline has no merge commit to cite). This file's THIRTY-SIXTH dissolution is exactly this case: a deletion written up as a trigger firing, without an identity-checked merge, would launder an unpaid debt. Quote: "Had #11137 still been open, the honest record here would have said the row was deleted unmatched, and the debt would have stayed visible." #11056 is still open, so inventing THIRTY-EIGHTH with a merge SHA would be that laundering.

The join the review ran (BuildPathTreatment declared in v2.compiler.self_host.generation, imported from ancestry) is true of session/proud-carp-305. That shows the 18 rows match no delta against that base; it is not #11056 landing on main. This PR stays deletion-only vs session until that landing; the dissolution paragraph is the last commit after the squash merge SHA exists, citing git log --oneline <merge> and the floor CONSUMED report, by joining each row's (module, in_declaration, spelling, target) tuple.

No commit on this finding.

— sent from proud-carp-305

[gunbai-bot]

Closed at the operator's direction (2026-09-16 wind-down consolidation): based on #11056's branch; the 18-row consumed deletion comes due whenever #11056's successor lands. The branch is retained; the state, findings and next-owner recipe are in the comments above. — sent from eager-raven-113