refactor(config): eliminate global mutable state, wire DeerFlowContext into runtime

- Freeze all config models (AppConfig + 15 sub-configs) with frozen=True
- Purify from_file() — remove 9 load_*_from_dict() side-effect calls
- Replace mtime/reload/push/pop machinery with single ContextVar + init_app_config()
- Delete 10 sub-module globals and their getters/setters/loaders
- Migrate 50+ consumers from get_*_config() to get_app_config().xxx

- Expand DeerFlowContext: app_config + thread_id + agent_name (frozen dataclass)
- Wire into Gateway runtime (worker.py) and DeerFlowClient via context= parameter
- Remove sandbox_id from runtime.context — flows through ThreadState.sandbox only
- Middleware/tools access runtime.context directly via Runtime[DeerFlowContext] generic
- resolve_context() retained at server entry points for LangGraph Server fallback

backend/packages/harness/deerflow/config/sandbox_config.py

@@ -4,6 +4,8 @@ from pydantic import BaseModel, ConfigDict, Field
 class VolumeMountConfig(BaseModel):
     """Configuration for a volume mount."""
 
+    model_config = ConfigDict(frozen=True)
+
     host_path: str = Field(..., description="Path on the host machine")
     container_path: str = Field(..., description="Path inside the container")
     read_only: bool = Field(default=False, description="Whether the mount is read-only")
@@ -80,4 +82,4 @@ class SandboxConfig(BaseModel):
         description="Maximum characters to keep from ls tool output. Output exceeding this limit is head-truncated. Set to 0 to disable truncation.",
     )
 
-    model_config = ConfigDict(extra="allow")
+    model_config = ConfigDict(extra="allow", frozen=True)