A check run records the base tip it was merged with, and the pull page and MCP name it #1585

Replaces #1178. Main already checks a pull merged into its base's tip (src/checks/merged.ts, service.ts ~L490, since 5d955dc8). What is missing is the record.

Done:

• [ ] A named migration adds base_oid to the check run; service.ts stores the tip it merged with. A push to the default branch records none.
• [ ] The pull page's checks line says which base it checked ("checked against main at abc1234").
• [ ] The MCP and the API's reading of a pull's check carry baseOid.
• [ ] A test fails without it: a pull's check run has the base tip it was merged with.

Files: src/checks/service.ts, a migration, src/web/pages/pull.ts, src/http/mcp.ts, test/checks.test.ts.

A test that goes red in this pull is fixed in this pull. Name any new test file in GATE in scripts/gate.ts.

Parked by the crew: the check failed and the decider saw no point in another try (pull #1628 closed). Planner: split this smaller, or rewrite what done looks like; a person can also say what to try.

Parked by the crew: main moved under it 3 times and the branch could not be brought up to date (src/core/schema.ts) (pull #1636 closed).

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

1. the check failed and the decider saw no point in another try (pull #1628 closed)
2. main moved under it 3 times and the branch could not be brought up to date (src/core/schema.ts) (pull #1636 closed)

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.

Deferred behind the Later gate (#595). The owner's direction of 2026-10-15 puts the check pipeline first, before any new feature: #1792, #1793, #1794, #1795, #1796, #1797, #1798, #1799, then #1789. Recording the base tip a check merged with overlaps #1795 (one green result per commit, no re-run on a moved base); it waits for that to land and is rewritten against it if still needed.

replaced by #1810. Correction to my note above: the Later gate refused it as a cycle (the gate waits on this pillar), so it is replaced instead, by an issue that waits for the owner's check-pipeline work it was failing on and that keeps its pillar.