Fix properly returning full conversation messages

app.py

@@ -127,12 +127,12 @@ def chat_with_model(messages):
 
         print(f'Step 2: {messages}')
 
-        yield messages[-1]["content"]
+        yield messages
 
     if in_think:
         output_text += "*"
         messages[-1]["content"] = output_text
-        yield messages[-1]["content"]
+        yield messages
 
     # Wait for thread to finish
     thread.join(timeout=1.0)
@@ -141,7 +141,7 @@ def chat_with_model(messages):
 
     print(f'Step 3: {messages}')
 
-    return messages[-1]["content"]
+    return messages