feat(US5+6): QA generation — POST /api/v1/qa/gen-text and /gen-image

- Add qa_models.py with TextQAItem, GenTextQARequest, QAPair, ImageQAItem,
  GenImageQARequest, ImageQAPair, TextQAResponse, ImageQAResponse
- Implement gen_text_qa(): batch-formats triples into a single prompt, calls
  llm.chat(), parses JSON array via extract_json()
- Implement gen_image_qa(): downloads cropped image from source-data bucket,
  base64-encodes inline (data URI), builds multimodal message, calls
  llm.chat_vision(), parses JSON; image_path preserved on ImageQAPair
- Replace qa.py stub with full router: POST /qa/gen-text and /qa/gen-image
  using Depends(get_llm_client) and Depends(get_storage_client)
- 15 new tests (8 service + 7 router), 53/53 total passing

app/routers/qa.py

@@ -1,3 +1,31 @@
-from fastapi import APIRouter
+from fastapi import APIRouter, Depends
+
+from app.clients.llm.base import LLMClient
+from app.clients.storage.base import StorageClient
+from app.core.dependencies import get_llm_client, get_storage_client
+from app.models.qa_models import (
+    GenImageQARequest,
+    GenTextQARequest,
+    ImageQAResponse,
+    TextQAResponse,
+)
+from app.services import qa_service
 
 router = APIRouter(tags=["QA"])
+
+
+@router.post("/qa/gen-text", response_model=TextQAResponse)
+async def gen_text_qa(
+    req: GenTextQARequest,
+    llm: LLMClient = Depends(get_llm_client),
+) -> TextQAResponse:
+    return await qa_service.gen_text_qa(req, llm)
+
+
+@router.post("/qa/gen-image", response_model=ImageQAResponse)
+async def gen_image_qa(
+    req: GenImageQARequest,
+    llm: LLMClient = Depends(get_llm_client),
+    storage: StorageClient = Depends(get_storage_client),
+) -> ImageQAResponse:
+    return await qa_service.gen_image_qa(req, llm, storage)