Gate completed checkpoints through training loop

docs/CODEX_EXECUTION_LOG.md

@@ -11777,6 +11777,11 @@ Deployment evidence:
   leak-free failure-driven sampler automatically, records its evidence
   checksum and next dataset YAML in `training-loop-state.json`, and resumes
   both the candidate weights and exact sampling input after interruption.
+- Added `--evaluate-initial-model` for completed checkpoints such as v37. It
+  skips redundant fitting only for the first iteration, copies and hashes the
+  supplied weights, runs calibration first, and rejoins the same automatic
+  sampling/training path after rejection. Protected evidence remains closed
+  until calibration passes.
 - Confirmed v37 epoch 1 completed on CUDA with validation precision `0.601`,
   recall `0.455`, mAP50 `0.474` and mAP50-95 `0.205`; the run remains inactive
   and these internal-validation metrics are not release evidence.
@@ -11786,7 +11791,7 @@ Verified in this pass:
 - `py -3 -m pytest -q backend/tests/test_belgium_training_loop.py backend/tests/test_belgium_training_iteration_assessment.py backend/tests/test_belgium_training_portfolio.py`
   (`12 passed`).
 - `py -3 -m pytest -q backend/tests/test_belgium_training_loop.py backend/tests/test_failure_driven_yolo_sampling.py backend/tests/test_belgium_training_iteration_assessment.py`
-  (`15 passed`).
+  (`16 passed` after adding the completed-checkpoint entry contract).
 
 Open: