ci(OMN-16589): pilot-wire the org-wide KB doc gate (transition mode)

[jonahgabriel]

What

Pilot-wires the org-wide KB doc gate (OmniNode-ai/omniclaude#2047, merged) into omnimarket: a thin caller workflow (.github/workflows/kb-doc-gate.yml) calling omniclaude's kb-doc-gate-reusable.yml@main, plus the exported kb-doc-gate pre-commit hook (repo: OmniNode-ai/omniclaude, pinned to the merged dev SHA).

Also fixes a real, live-caught CI failure: test_every_pr_triggered_job_is_classified failed because the new kb-doc-gate.yml PR-triggered job wasn't classified anywhere in the CI Summary poller's registry. Registered it in EXEMPT_CONTEXTS (self-declared non-required, same pattern as product-readiness-shadow.yml's Phase-3-canary entries) — NOT EXPECTED_EXTERNAL_CONTEXTS, since that would make CI Summary treat it as de facto required, contradicting this ticket's P1 scope.

Why

Operator direction (2026-08-26): a hook that blocks new/regrown non-KB markdown docs in product repos. See omniclaude#2047 for the full design rationale.

Scope

NOT a required status check yet. Transition mode: blocks new non-exempt .md files only; the strict-mode stub-size cap is off since this repo has not yet been stripped by the Wave-4 KB migration. The required-check flip is explicit follow-up scope once this pilot (omnibase_core + omnimarket) proves green on real PRs — see OMN-16589 AC.

Testing

uv run pytest tests/unit/scripts/ci/test_ci_summary_gate.py -q → 41 passed. Governed pre-push impacted-test selector: full local suite (5830 passed, 29 skipped, 1 xfailed) green under PREPUSH_ALLOW_LOCAL_FULL_SUITE=1 — used because the shared build host and the .201 gate-runner were both confirmed at/over the load threshold by the hook's own live check at push time (session-wide contention from ~70 concurrently active lanes), not as a routine bypass; see docs/runbooks/200-build-lane-execution-pattern.md.

OMN-16589

Evidence-Ticket: OMN-16589
Evidence-Source: OCC#7185

[coderabbitai]

Warning

Review limit reached

Next included review available in 28 minutes.

View limit details

Limit details: You’ve used the included review currently available. Your 124 included PR review attempts over the past 7 days set your current allowance at 1 review per hour.

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 49e845e5-1ec3-483b-9ac2-ce937f3a0e5b

📥 Commits

Reviewing files that changed from the base of the PR and between fcc8c58 and 3db1d52.

📒 Files selected for processing (3)

• .github/workflows/kb-doc-gate.yml
• .pre-commit-config.yaml
• tests/unit/scripts/ci/test_ci_summary_gate.py

---

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

✅ Architectural Review — PASSED

Errors: 0
Warnings: 0

What this checks

Rule	Description
ARCH-TOPIC-001	No hardcoded Kafka topic strings in handler code
ARCH-DI-001	No event_bus=None bypass
ARCH-DI-002	No direct Handler instantiation outside workflow_runner/adapters
ARCH-DI-003	No reinvented DI containers
ARCH-TOPIC-002	contract.yaml topics follow `onex.{cmd

---

No architectural violations found.

Static architectural lint — no model inference (OMN-14176).

[onexbot-occ-writer]

OCC autobind did not mint a companion for this PR: no changed-file candidate could be proven RED against the merge base, and emitting a PR-existence probe instead would be non-falsifiable evidence (OMN-15247). Hand-authored evidence is required.

[github-actions]

✅ Hostile Reviewer — PASSED

Blocking findings (critical/error): 0
Total findings: 0
Models succeeded: qwen3-review,qwen3-review-b

---

Gate semantics

Verdict	Meaning	Blocks merge?
passed	>=2 models succeeded, no critical/error findings	No
blocked	CRITICAL or ERROR findings found	Yes
degraded	Fewer than 2 models succeeded (infra unavailable/timeout)	Yes (OMN-15110)

Powered by omniintelligence.review_pairing.cli_review — node-based adversarial review via HandlerLlmCliSubprocess (OMN-8524)