Add Chatgpt Batch Detail Fetch

Why

The ChatGPT connector currently hydrates conversation detail with one GET /conversation/{id} request per conversation. Large accounts turn that into a provider request storm, increasing run time and 429 pressure even though ChatGPT exposes a batch detail endpoint that returns the same conversation-detail shape for up to 10 ids per request.

Reducing the detail-fetch request count improves connector reliability without changing emitted records, cursor shape, grant semantics, or owner-facing data.

What Changes

• Add a ChatGPT API seam for POST /conversations/batch with the same retry helper used by existing requests.
• Hydrate listed conversation ids in batches of at most 10 before the existing per-conversation processing lane.
• Preserve the current GET /conversation/{id} path as fallback for ids omitted by the batch response or when the batch endpoint is unavailable.
• Add regression tests proving batch-hit, batch-omission, batch-cap, and fallback behavior.

Capabilities

Modified:

• polyfill-runtime

Impact

• Affected code: ChatGPT connector API wrapper, ChatGPT detail-hydration orchestration, ChatGPT integration tests, connector version note.
• Validation: targeted ChatGPT connector tests, connector typecheck where available, OpenSpec strict validation.
• Risk: the batch endpoint is an undocumented provider endpoint. The implementation must fail open to the existing per-id GET path so collection remains correct if batch is unavailable or omits records.