Default retrieval to 6 chunks instead of 2

chunks is a total per request, split across the attached sources, so an
agent with two sources and the default of 2 got a single chunk from each,
and its answers changed with whichever chunk won. 6 gives three per
source for about 4-5k more input tokens per retrieval turn.

Every literal default moves from 2 to 6: the request default, the
retrievers, the internal search tool, workflow agent nodes, scheduled and
headless runs, agent create/update/import, the source retrieval config and
the frontend forms. Existing agents and sources keep what they store; a
source saved with chunks=2 now counts as configured at 2, which is pinned
by a test.

Headless runs also read chunks=0 as unset (`or 2`), so an agent with
retrieval switched off retrieved anyway on scheduled runs; 0 now stays 0.

docsgpt/storage/db/source_config.py

@@ -118,7 +118,7 @@ class RetrievalConfig(BaseModel):
 
     retriever: str = "classic"  # RetrieverCreator key
     exposure: str = "prefetch"  # prefetch | agentic_tool (D11)
-    chunks: int = 2  # final top-k
+    chunks: int = 6  # final top-k
     score_threshold: Optional[float] = None  # pgvector/mongo honor it; others ignore
     rephrase_query: bool = True  # toggle ClassicRAG._rephrase_query side-call
     reranker: Optional[dict] = None  # reserved: future cross-encoder/LLM reorder