Add request_id and turn_id as log context, clearly indicate model inference errors

[chiragjn]

Note

Low Risk
Observability and error-message changes only; no auth or persistence behavior changes beyond safer error logging.

Overview
Request-scoped logging adds request_id (from x-request-id or minted) on every HTTP request via AsyncLocalStorage and Winston, echoes it on responses, binds turn_id when a turn starts, and propagates request_id from schedule dispatch through the controller’s internal API client. The version field is removed from per-line JSON logs.

Model and error logging wraps LLM failures as Model request failed: <provider>/<model>: <reason> (aborts unchanged), deduplicates model errors so only AgentThread logs once, and tightens extractErrorLogFields: chained cause stacks, no JSON dumps that could leak SDK request bodies (prompts).

Tests cover toStreamError, cause stacks, and prompt redaction.

^{Reviewed by Cursor Bugbot for commit e0adf15. Bugbot is set up for automated code reviews on this repo. Configure here.}

[changeset-bot]

🦋 Changeset detected

Latest commit: e0adf15

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 2 packages

Name	Type
@truefoundry/trueforge-core	Patch
@truefoundry/trueforge	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR