Streamlit Stream Callback in Haystack 2.x

Describe the solution you'd like
I would like to see streaming response in real time on streamlit when using OpenAIChatGenerator

Describe alternatives you've considered
I have tried to the following streaming_callback function. (I have only put related code fields. I can show full code if requested)

import streamlit as st

# Method 1: Nothing seen in the UI
def st_streaming_callback(chunk: StreamingChunk):
    yield chunk.content

# Method 2: Instead of typewriter format, each chunk seen as one line output.
def st_streaming_callback(chunk: StreamingChunk):
    st.write(chunk.content)

model = OpenAIChatGenerator(
    model="gpt-4-turbo-preview",
    streaming_callback=streaming_callback,
    generation_kwargs=None,
)

response_dict = st.write_stream(
  llm_chat.run(
      messages=[
          ChatMessage(
              role=message.role, content=message.content, name=None
          )
          for message in st.session_state.messages
      ]
 )

Additional context
Add any other context or screenshots about the feature request here.