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/deer_flow_context.py

@@ -0,0 +1,59 @@
+"""Per-invocation context for DeerFlow agent execution.
+
+Injected via LangGraph Runtime. Middleware and tools access this
+via Runtime[DeerFlowContext] parameters, through resolve_context().
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any
+
+
+@dataclass(frozen=True)
+class DeerFlowContext:
+    """Typed, immutable, per-invocation context injected via LangGraph Runtime.
+
+    Fields are all known at run start and never change during execution.
+    Mutable runtime state (e.g. sandbox_id) flows through ThreadState, not here.
+    """
+
+    app_config: Any  # AppConfig — typed as Any to avoid circular import at module level
+    thread_id: str
+    agent_name: str | None = None
+
+
+def resolve_context(runtime: Any) -> DeerFlowContext:
+    """Extract or construct DeerFlowContext from runtime.
+
+    Gateway/Client paths: runtime.context is already DeerFlowContext → return directly.
+    LangGraph Server / legacy dict path: construct from dict context or configurable fallback.
+    """
+    ctx = getattr(runtime, "context", None)
+    if isinstance(ctx, DeerFlowContext):
+        return ctx
+
+    from deerflow.config.app_config import AppConfig
+
+    # Try dict context first (legacy path, tests), then configurable
+    if isinstance(ctx, dict):
+        return DeerFlowContext(
+            app_config=AppConfig.current(),
+            thread_id=ctx.get("thread_id", ""),
+            agent_name=ctx.get("agent_name"),
+        )
+
+    # No context at all — fall back to LangGraph configurable
+    try:
+        from langgraph.config import get_config
+
+        cfg = get_config().get("configurable", {})
+    except RuntimeError:
+        # Outside runnable context (e.g. unit tests)
+        cfg = {}
+
+    return DeerFlowContext(
+        app_config=AppConfig.current(),
+        thread_id=cfg.get("thread_id", ""),
+        agent_name=cfg.get("agent_name"),
+    )