experiment(confidence-engine): complete openai reconstruction apparatus

docs/current-handoff.md

@@ -134,6 +134,17 @@
 - Its import-only proof succeeded with `startCaseResolved: true`: the real `startCase()` dependency chain resolves and no provider call occurs. Production runtime and the Ollama/Qwen default remain unchanged.
 - Next boundary: exactly one live GPT-5.6 Terra reconstruction through this canonical helper.
 
+## OpenAI CLI experiment selection
+
+- The canonical helper now exposes experiment-only OpenAI selection through `START_CASE_EXPERIMENT_PROVIDER=openai`, reusing the existing OpenAI provider constructor and provider-injection seam. Its default behavior is unchanged.
+- Import-only mode remains inert, production provider selection remains Ollama/Qwen, and no live calls occurred. Next boundary: exactly one live OpenAI call using the fixed manufacturing scenario.
+
+## OpenAI strict reconstruction transport apparatus
+
+- The first real OpenAI request reached Responses API but failed before inference with `invalid_json_schema`; the first reported incompatibility was optional `unexplainedTransitions[].entity` absent from `required`.
+- OpenAI strict transport projection now derives structure from canonical Zod-generated JSON Schema and canonical input optionality from the Zod contract, including `.optional().default(...)`. Canonically omittable fields are required-but-nullable for transport; null placeholders are omitted only for those fields before canonical Zod validation, while genuinely required fields remain non-nullable.
+- `reconstructionV2Schema` and Ollama behavior remain unchanged. `START_CASE_EXPERIMENT_PROVIDER=openai` is experiment-only selection; default behavior remains Ollama/Qwen. Deterministic closeout passed with zero live calls; next boundary is exactly one live GPT-5.6 Terra reconstruction.
+
 ## Current product architecture
 
 Three distinct routes, not a single page: