feat/ updated the ```generate_llm_response``` method for it to stream the text as the model generates it.

app.py

@@ -123,11 +123,10 @@ def generate_llm_response(
     t.start()
     
     # Collect the output
-    outputs = []
+    accumulated_response = ""
     for text in streamer:
-        outputs.append(text)
-    
-    return "".join(outputs)
+        accumulated_response += text
+        yield accumulated_response
 
 def append_text_knowledge(file_path: str) -> str:
     """