fix: stop empty hops from draining the pool and sync gpt-6 models

[dofastted]

Summary

• An empty or thinking-only hop retries the same account once and returns 502 incomplete_response. That request no longer parks or rotates the rest of the pool. A later request that empty-hops the same account parks only that account for 60 seconds.
• Oversized inbound bodies fail with 413 body_too_large before scheduling. The default limit is 128MB (KIN_MAX_BODY still overrides it).
• Live Codex account sync now includes gpt-6-sol and gpt-6-luna.

Test plan

• node --test test/unit/failover-runner.test.mjs test/unit/rate-limit-service.test.mjs test/unit/respond-sse.test.mjs
• Confirm a single incomplete hop leaves other accounts schedulable
• Confirm a body over 128MB returns 413 and does not select an account

Made with Cursor

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_6e43661f-6975-43e3-a6a5-97804d3e1e7a)