refactor(config): thread AppConfig through RunContext into Worker

Phase 2 Task P2-5 (partial, Category D): RunContext gains an optional
app_config field. Gateway deps.get_run_context populates it from
app.state.config. Worker builds DeerFlowContext from ctx.app_config when
provided, falling back to AppConfig.current() for legacy callers.

This is the key path: Gateway-mode agent runs now construct
DeerFlowContext without consulting the process-global at all — app_config
flows from FastAPI's app.state through RunContext to Worker to
Runtime.context. Standalone providers (checkpointer/store/stream_bridge)
already accept optional config parameter from P1, so they are Phase 2-ready.

59 gateway/worker/run_manager tests pass.

backend/app/gateway/deps.py

@@ -149,6 +149,7 @@ def get_run_context(request: Request) -> RunContext:
         event_store=get_run_event_store(request),
         run_events_config=getattr(config, "run_events", None),
         thread_store=get_thread_store(request),
+        app_config=config,
     )
 
 

backend/packages/harness/deerflow/runtime/runs/worker.py

@@ -54,6 +54,9 @@ class RunContext:
     run_events_config: Any | None = field(default=None)
     thread_store: Any | None = field(default=None)
     follow_up_to_run_id: str | None = field(default=None)
+    # Phase 2: app-level config flows through RunContext so Worker can build
+    # DeerFlowContext without consulting the process-global.
+    app_config: Any | None = field(default=None)
 
 
 async def run_agent(
@@ -169,7 +172,7 @@ async def run_agent(
         # LangGraph's astream(context=...) injects this into Runtime.context
         # so middleware/tools can access it via resolve_context().
         deer_flow_context = DeerFlowContext(
-            app_config=AppConfig.current(),
+            app_config=ctx.app_config if ctx.app_config is not None else AppConfig.current(),
             thread_id=thread_id,
         )