Refactor csc serve question endpoints to match opencode/app-ai-native
expected data models and request/response shapes:
- GET /question now returns QuestionRequest[] (direct array) instead of
{ questions: [...] }
- POST /question/:id/reply accepts { answers: string[][] } mapping to
selected option labels per question
- POST /question/:id/reject no longer requires a request body
- Support dual question sources:
- MCP Elicitation (converted to opencode format with enum→options)
- AskUserQuestionTool permissions exposed as questions
- Add conversion helpers: convertAskUserQuestionToOpencode,
convertElicitationToOpencode, buildElicitationContent
- Reply to AskUserQuestionTool via replyPermission('allow') with
constructed updatedInput.answers map
Also includes serve-mode infrastructure improvements:
- Session routes auto-create sessions on demand via getOrCreateSession
- Set CSC_SERVE_MODE env var in child process spawn
- Skip push-suggestion stream waiting in serve mode
Add comprehensive unit tests for question route opencode conversion.
Co-authored-by: CoStrict <zgsm@sangfor.com.cn>
QueryEngine previously guarded stream_event yields behind includePartialMessages,
which was undefined in serve mode. Remove the guard so stream_event messages
always flow to the child process stdout. Add stream_event handling in
sessionHandle.routeMessage to emit via EventBus for downstream consumers.
- Add _providerId field to SessionHandle, extracted from initData.account.apiProvider
- Include provider_id in user message emit, assistant message fallback, and session.ready
- Revert getProviderId normalization (keep raw value from subprocess)
The subprocess returns apiProvider='firstParty' for direct Anthropic access,
but the adapter hardcodes providerID='anthropic' in SSE events. This mismatch
causes the frontend's isModelValid check to fail, clearing the model selector.
- Remove process.stderr.write debug logs from sessionHandle and prompt_async
- Add model fallback in assistant message (use this._model if msg.model missing)
- Add model field to user message emit for adapter consumption
- Wait for session ready before setModel/prompt in prompt_async
- Return 200 JSON instead of 204 No Content from prompt_async
- Set Bun.serve idleTimeout to 255s to prevent connection drops
