Record promoted detector runtime verification

CHANGELOG.md

@@ -14,6 +14,7 @@
 - Compared fixed-threshold persisted evidence against the previous active model and reduced the false-negative rate in every validated positive AOI.
 - Activated the exact promoted local model through the guarded dry-run-first helper; no model was downloaded and no fake inference or QA result was introduced.
 - Updated Detection Lab operator profiles so the promoted expanded model is the recommended balanced review choice while the previous high-precision model remains available as a legacy conservative profile.
+- Redeployed the all-in-one Tower runtime with CPU-only Torch `2.13.0+cpu`, verified a real local model load against a nine-tile manifest, and passed PostGIS, API, health and browser checks.
 - Persistent small-building misses remain the primary model-quality limitation and still require operator QA/QC.
 
 ## Sprint 172 CPU AI image build hardening (2026-07-12)

docs/CODEX_EXECUTION_LOG.md

@@ -7061,3 +7061,13 @@ Open:
 ## Remaining limitation
 
 - Persistent small-building misses remain the weakest quality bucket. Continue with targeted evidence review and data coverage before considering another training run; do not infer production accuracy from aggregate F1 alone.
+
+## Tower deployment verification
+
+- Pushed commit `1c16313` and rebuilt the all-in-one runtime at `http://192.168.10.150:1202`.
+- The opt-in AI image installed `torch 2.13.0+cpu` and `torchvision 0.28.0+cpu` from the official CPU wheel index; runtime CUDA availability is `false`. Ultralytics reports version `8.4.93`.
+- The live model catalog exposes 23 local files and marks only `geointel-building-yolov8s-aoi1024expandedminpx4vis035e50-pt` active with the expected SHA256.
+- Explicit `--check-model-load` preflight passed against an existing nine-tile manifest: dependencies, model load, manifest structure, tile paths and tile limit were all valid; no inference or download ran.
+- Embedded PostGIS live migration smoke passed with PostGIS `3.6`, required tables/indexes, database collation and the single Alembic head `202606120900`.
+- Browser validation confirmed that the recommended profile selects the exact active asset and threshold `0.15`, while live preflight displays CPU dependency/model readiness and the expected missing-manifest guard before dataset handoff.
+- Browser console warnings/errors: `0`.

docs/TODO.md

@@ -136,7 +136,7 @@ This file now starts with the current implementation status. Older preparation/b
 - [x] Train one inactive candidate from the filtered AOI1024 cleanpx YOLO dataset and gate it through the positive-AOI plus split-background promotion workflow; reject it because mean positive F1 remains below gate.
 - [x] Add deterministic dataset/base/trained-model SHA256 provenance to future operator training summaries.
 - [x] Review per-AOI false-negative evidence, expand positive sample/label coverage and verify the resulting candidate improves false-negative rate in every validated AOI.
-- [ ] Complete rebuild/restart and browser/runtime smoke for the guarded promoted V1 building detector activation.
+- [x] Complete rebuild/restart and browser/runtime smoke for the guarded promoted V1 building detector activation.
 - [ ] Expand focused small-building training evidence only after reviewing persistent false negatives from the promoted model; do not start another blind training run.
 
 ## Sprint 8 status