fix: handle non-picklable original_exception in MidStreamFallbackError

The __getstate__ mixin was storing attributes as-is into the pickle
state dict. For openai SDK exceptions (e.g. openai.RateLimitError),
pickle.dumps succeeds but pickle.loads fails because the openai
__init__ requires response and body keyword arguments that the
default pickle reconstruction does not supply.

Fix: probe each non-httpx attribute with a full round-trip
(pickle.loads(pickle.dumps(v))) before storing it. If the round-trip
fails, store str(v) instead so the attribute is still meaningful
after unpickling.

Covers MidStreamFallbackError.original_exception and any other
attribute that is dumps-able but not loads-able.

litellm/exceptions.py

@@ -9,6 +9,7 @@
 
 ## LiteLLM versions of the OpenAI Exception Types
 
+import pickle
 from typing import Optional
 
 import httpx
@@ -87,7 +88,11 @@ class _LiteLLMPickleMixin:
             elif isinstance(v, httpx.Request):
                 http_attrs.append(k)
             else:
-                state[k] = v
+                try:
+                    pickle.loads(pickle.dumps(v))
+                    state[k] = v
+                except Exception:
+                    state[k] = str(v)
         if http_attrs:
             state["_pickled_http_attrs"] = http_attrs
         return state