fix zhipuAI stream issue (#825)

### What problem does this PR solve?


### Type of change

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

api/apps/api_app.py

             resp.headers.add_header("Content-Type", "text/event-stream; charset=utf-8")
             return resp
         else:
-            ans = chat(dia, msg, False, **req)
+            ans = chat(dia, msg, **req)
             fillin_conv(ans)
             API4ConversationService.append_message(conv.id, conv.to_dict())
             return get_json_result(data=ans)

api/apps/conversation_app.py

             return resp
 
         else:
-            ans = chat(dia, msg, False, **req)
+            ans = chat(dia, msg, **req)
             fillin_conv(ans)
             ConversationService.update_by_id(conv.id, conv.to_dict())
             return get_json_result(data=ans)

rag/llm/chat_model.py

                 if not resp.choices[0].delta.content:continue
                 delta = resp.choices[0].delta.content
                 ans += delta
-                tk_count = resp.usage.total_tokens if response.usage else 0
-                if resp.output.choices[0].finish_reason == "length":
+                if resp.choices[0].finish_reason == "length":
                     ans += "...\nFor the content length reason, it stopped, continue?" if is_english(
                         [ans]) else "······\n由于长度的原因，回答被截断了，要继续吗？"
+                    tk_count = resp.usage.total_tokens
+                if resp.choices[0].finish_reason == "stop": tk_count = resp.usage.total_tokens
                 yield ans
         except Exception as e:
             yield ans + "\n**ERROR**: " + str(e)