Fix hit cache error while raptoring. (#4955)

### What problem does this PR solve?

#4126

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

rag/raptor.py

@@ -46,7 +46,7 @@ class RecursiveAbstractiveProcessing4TreeOrganizedRetrieval:
 
     def _embedding_encode(self, txt):
         response = get_embed_cache(self._embd_model.llm_name, txt)
-        if response:
+        if response is not None:
             return response
         embds, _ = self._embd_model.encode([txt])
         if len(embds) < 1 or len(embds[0]) < 1: