Fix: fix error 429 api rate limit when building knowledge graph for all chat model and Mistral embedding model (#9106)

### What problem does this PR solve?

fix error 429 api rate limit when building knowledge graph for all chat
model and Mistral embedding model.

### Type of change

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

rag/llm/chat_model.py

@@ -73,7 +73,7 @@ class Base(ABC):
 
     def _get_delay(self):
         """Calculate retry delay time"""
-        return self.base_delay + random.uniform(10, 150)
+        return self.base_delay + random.uniform(60, 150)
 
     def _classify_error(self, error):
         """Classify error based on error message content"""

rag/llm/embedding_model.py

@@ -463,25 +463,42 @@ class MistralEmbed(Base):
         self.model_name = model_name
 
     def encode(self, texts: list):
+        import time
+        import random
         texts = [truncate(t, 8196) for t in texts]
         batch_size = 16
         ress = []
         token_count = 0
         for i in range(0, len(texts), batch_size):
-            res = self.client.embeddings(input=texts[i : i + batch_size], model=self.model_name)
+            retry_max = 5
-            try:
+            while retry_max > 0:
-                ress.extend([d.embedding for d in res.data])
+                try:
-                token_count += self.total_token_count(res)
+                    res = self.client.embeddings(input=texts[i : i + batch_size], model=self.model_name)
-            except Exception as _e:
+                    ress.extend([d.embedding for d in res.data])
-                log_exception(_e, res)
+                    token_count += self.total_token_count(res)
+                    break
+                except Exception as _e:
+                    if retry_max == 1:
+                        log_exception(_e)
+                    delay = random.uniform(20, 60)
+                    time.sleep(delay)
+                    retry_max -= 1
         return np.array(ress), token_count
 
     def encode_queries(self, text):
-        res = self.client.embeddings(input=[truncate(text, 8196)], model=self.model_name)
+        import time
-        try:
+        import random
-            return np.array(res.data[0].embedding), self.total_token_count(res)
+        retry_max = 5
-        except Exception as _e:
+        while retry_max > 0:
-            log_exception(_e, res)
+            try:
+                res = self.client.embeddings(input=[truncate(text, 8196)], model=self.model_name)
+                return np.array(res.data[0].embedding), self.total_token_count(res)
+            except Exception as _e:
+                if retry_max == 1:
+                    log_exception(_e)
+                delay = random.randint(20, 60)
+                time.sleep(delay)
+                retry_max -= 1
 
 
 class BedrockEmbed(Base):