feat(OMN-17985): boot the seven projection-writer profiles core 0.47.5 registered

[jonahgabriel]

OMN-17985 (AC2, behaviour half) — boot the seven profiles core 0.47.5 registered

omnibase_core 0.47.5 (#1656, published to PyPI) admits projection-writer-{delegation,hook-ledger,live-events,registration,savings,tenant-credentials,tenant-registry} into REGISTERED_RUNTIME_PROFILES and CONSUMER_ATTACHED_RUNTIME_PROFILES. Core owns the names; this repo owns the behaviour. Two guards here make the pairing mandatory rather than cosmetic:

• test_profiles_match_core_registry asserts frozenset(_PROFILES) == REGISTERED_RUNTIME_PROFILES — exact set equality, both directions.
• test_consumer_attached_profiles_actually_load requires every consumer-attached name to resolve through load_runtime_profile.

Why this is now a hard failure rather than a silent one

#3238 made load_runtime_profile refuse an unknown name. So a name core blesses that this dict lacks is no longer a warning-plus-fallback-to-default — it is a boot failure. The RED run demonstrates exactly that: bumping the core pin to 0.47.5 with _PROFILES unchanged failed both guards, the second with

ProtocolConfigurationError: Unknown RUNTIME_PROFILE 'projection-writer-live-events'

which is precisely the core/infra drift the parity guard was written for, finally loud instead of silent.

prefetch_policy stays "disabled" on all seven, like every other role-based profile.

No lane's behaviour changes

These names are already the deployed RUNTIME_PROFILE values on onex-dev, and the processes carrying them are standalone BaseProjectionRunner runners that never read the variable. Verified live after this change was authored: probe run 34038648216 shows all seven 1/1 Ready with broker-issued partition assignments, public.live_events at 293,555 rows and growing.

Companion surfaces rebound in the second commit

Every commit that moves pyproject.toml or uv.lock rebinds three mirrors of the core pin. #3242 made the identical companion edit for the 0.47.3 → 0.47.4 move:

• tests/fixtures/seams/core_release/0.46.11_expected_symbols.json — freezes the released core version.
• src/omnibase_infra/runtime/version_compatibility.py — min_version floor.
• docker/runners/runner-image.lock.json — identity_digest + shared_env_digest, regenerated with scripts/ci/runner_image_identity.py --mode generate and then re-verified.

The first two were caught as genuine reds by the governed pre-push full-suite escalation on h201.2 (2 failed / 26,733 passed) and are fixed here rather than waived. The third was found by running the verifier rather than waiting for CI to report it.

Verification

• uv run pytest tests/unit/runtime/test_seam_released_core_pin.py tests/unit/runtime/test_fallback_matrix_sync.py tests/unit/runtime/test_runtime_profile.py tests/unit/runtime/test_runtime_profile_fail_closed.py => 60 passed.
• uv run pytest tests/ci/test_lane_projection_writer_coverage_omn17562.py tests/integration/test_writer_owned_projection_wiring_omn17519.py testsets/unit/runtime/test_kernel.py => 71 passed, 3 skipped.
• uv run python scripts/ci/runner_image_identity.py --mode verify => runner image identity verified: v7 dd687c47e140a3a8680114ad3fba9103.
• ruff format / ruff check / mypy --strict => clean. pre-commit on the changed paths => all pass.

0.38.20 is already the version on dev and 0.38.19 is the latest published, so this rides the existing unreleased bump; no further version move is needed.

DoD evidence: OMN-17985, acceptance criterion 2 (behaviour half) and criterion 4 (red-then-green).

Evidence-Ticket: OMN-17985
Evidence-Source: OCC#8438

[github-actions]

Hostile Reviewer — adversarial findings (OMN-17492)

Models succeeded: glm-review
Models failed: codex
New finding threads: 5
Deduped (already posted on this PR): 0
Nit-level findings suppressed: 0

The model is the FINDER, never the gate: merge is gated only by the
deterministic Hostile Review Thread Gate, which blocks while
hostile-reviewer threads are unresolved. Resolve each thread after
addressing (or rejecting, with a reply) its finding.

Findings demoted from threads (anchor rejected)

• [MAJOR] hostile-reviewer (glm-review)

  Seam fixture version bumped without recording new 0.47.5 surface | The fixture's stated purpose is to fail loudly on the exact symbols and field types consumed from core, with an update procedure of 'bump core, re-run the seam test, and record the new surface'. The diff changes core_version to 0.47.5 but the modules block is byte-identical. Core 0.47.4 was described as a pre-release pulling merged PRs #1439 and #1440; a released 0.47.5 may diverge from that snapshot, yet no re-derivation of the expected sym

  Resolve this thread when addressed — the Hostile Review Thread Gate blocks while hostile-reviewer threads are unresolved (OMN-17492).

• [MAJOR] hostile-reviewer (glm-review)

  Min-version bump in fallback matrix breaks running 0.47.4 deployments on restart | _FALLBACK_MATRIX now enforces min_version 0.47.5. Any environment still running omnibase-core 0.47.4 (e.g., a pod using a stale cached image during a rolling deploy, since the runner image lock is only updated in the same merge) will fail the compatibility gate on next process start. The diff updates the pin everywhere atomically but provides no transition window or explicit allowlist for the outgoing version, converting a ro

  Resolve this thread when addressed — the Hostile Review Thread Gate blocks while hostile-reviewer threads are unresolved (OMN-17492).

• [MINOR] hostile-reviewer (glm-review)

  Seven near-identical profile literals instead of a factory | The seven projection-writer profiles differ only in name. The comment explains why they must exist locally, but the repetition is unnecessary and each future profile addition requires seven lines of boilerplate plus exact-match parity. A comprehension generating them from a name tuple would be equivalent under the existing set-equality tests and reduce the chance of a copy-paste typo in name versus dict key, which current tests may not distinguish

  Resolve this thread when addressed — the Hostile Review Thread Gate blocks while hostile-reviewer threads are unresolved (OMN-17492).

• [MINOR] hostile-reviewer (glm-review)

  Comment asserts ownership claim, no test enforces it | The comment states the profile names 'settle' ownership of the seven writers and that previously contracts were double-claimed by shared runtimes. The diff adds no test that a shared runtime profile (effects, main) fails to load if it attempts to attach these writers, nor that KAFKA_CONSUMER_GROUP values are distinct. The parity tests cited only verify name-set equality and loadability; the actual ownership invariant is documented prose, not an enforced

  Resolve this thread when addressed — the Hostile Review Thread Gate blocks while hostile-reviewer threads are unresolved (OMN-17492).

• [MINOR] hostile-reviewer (glm-review)

  Runner image lock digests updated without provenance | identity_digest and shared_env_digest change alongside the omnibase-core pin, which is expected, but the diff contains no evidence the image was rebuilt from the new uv.lock rather than hand-edited. If the digests were computed against a stale lock or a different base, the runner image and the Python environment silently diverge; the lock file gives no way to verify. | Evidence: - "identity_digest": "b897b566327774f10e9f0c7208c49e2f",

  • "identity_dige

  Resolve this thread when addressed — the Hostile Review Thread Gate blocks while hostile-reviewer threads are unresolved (OMN-17492).

[github-actions]

⚠️ Hostile Reviewer — DEGRADED (informational)

Critical findings: 0
Major findings: 2
Total findings: 5
Models succeeded: glm-review

Note: Fewer than 2 reviewer models succeeded. Degraded results are informational (OMN-8468/OMN-8524) and do not block merge. Error: cli_review exit 2 (fewer than 2 models succeeded — partial/total outage)

---

Semantics (OMN-17492 — the model finds, thread resolution gates)

Surface	Meaning	Blocks merge?
Review threads	Per-finding, posted by the reviewer	No (informational)
Hostile Review Thread Gate	Deterministic: unresolved hostile-reviewer threads exist	Fails until resolved (not yet a required context)
degraded verdict	Fewer than 2 models succeeded (infra)	No

Powered by omniintelligence.review_pairing.cli_review — multi-model adversarial review: qwen3-review, qwen3-review-b, glm-review (OMN-8468/OMN-8524/OMN-17492)