修正response使用方法

gpt4.py

@@ -27,7 +27,7 @@ client = OpenAI(api_key=OPENAI_API_KEY)
 text_system_prompt = "你是一個中文的AI助手，請用繁體中文回答"
 
 # === 先建立第一個對話，之後可以延續這個對話 ===
-init_response = client.responses.create(
+response = client.responses.create(
     model="gpt-4o-mini",
     input=[{"role": "system", "content": text_system_prompt}],
 )
@@ -54,12 +54,12 @@ handler = WebhookHandler(channel_secret)
 
 # === AI Query 包裝 ===
 def query(payload):
-    response = client.responses.create(
+    second_response = client.responses.create(
         model="gpt-4o-mini",
-        previous_response_id=init_response.id,
+        previous_response_id=response.id,
         input=[{"role": "user", "content": f"{payload}"}],
     )
-    return response.output_text
+    return second_response.output_text
 
 
 # === 靜態圖檔路由 ===