docs(register): C-340 and C-341 — two findings that fail green

[Polichinel]

/register-risk on the v1.11.0 close-out. Both findings were acted on during the session and never tracked.

C-340 — Auto-merge fails silently, two mechanisms (Tier 3)

1. The method silently does not change. gh pr merge --auto --<method> refuses to change the method on an already-armed PR and reports nothing. During v1.10.0 that left squash armed on a development → main promotion. A squash there rewrites the release SHAs and permanently breaks the ancestry the back-merge exists to maintain — and it would not have surfaced until a later release diffed against a base that never existed. Caught only by reading auto_merge.merge_method back.

2. Pushing to an already-merged branch orphans the work. #416 merged the instant CI went green; the next push landed on a branch with no open PR, and two pieces of work were simply not on development. git push reported success. The only signal was commits=1 contradicting a remembered second push.

Tier 3, not 2: no data or model-output impact, both recoverable. Not Tier 4: auto-merge is armed on every PR here, so both are live on every change, and mechanism 1's damage to main would be permanent.

C-341 — Deploy gates run only where someone types pytest (Tier 4)

The unexamined residual of C-320's own fix. Making gates skip-with-reason where CI cannot answer was right; the consequence — that they assure only whoever runs the suite at the right moment — was never looked at. Cost four hours of undetected divergence after v1.10.0, and had been silently true after every prior release. Partially mitigated by release-topology.yml; the rest remain local-only.

Why these two, together

Both fail green. git push succeeds; gh pr merge exits 0; a skipped test is not a red test. Same class as C-330's nightly no-op reporting success and C-337's lockfile frozen without an error — which is the pattern this whole week produced.

Skipped after dedup

Finding	Reason
numpy/pyarrow/zarr floors vs. their locks	Duplicate — already inside C-337 (:566) as its open residual
views-faoapi monitor observations	Different repo's register
gh 2.4.0 lacking flags	Resolved; residual is the operator's machine

Verification

339 → 341 IDs · 43 → 45 open · 0+2+12+25+6 = 45 · struck-through 109 unchanged · index 1723 < 8000 · header 1723 ≤ 3500 · all 13 register guards + 7 changelog guards green · full suite exit 0 · validate_docs.sh exit 0