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.

tests/api/answer/test_stream_processor.py

@@ -1202,7 +1202,7 @@ class TestConfigureAgent:
         assert sp.retriever_config["chunks"] == 5
 
     @pytest.mark.unit
-    def test_configure_agent_invalid_chunks_defaults_to_2(self):
+    def test_configure_agent_invalid_chunks_uses_the_default(self):
         sp = self._make_sp()
         sp._resolve_agent_id = MagicMock(return_value="agent_id_1")
         sp._get_agent_key = MagicMock(return_value=("agent_key", False, None))
@@ -1220,7 +1220,7 @@ class TestConfigureAgent:
         sp._configure_agent()
         sp.model_id = "test-model"
         sp._configure_retriever()
-        assert sp.retriever_config["chunks"] == 2
+        assert sp.retriever_config["chunks"] == 6
 
 
 # ---- Additional coverage: _load_conversation_history ----