feat(chat): add message retry functionality

- [feat] Add `handle_chat_retry` function to implement assistant response retry logic (chat_handler.py:196)
- [feat] Implement retry logic in `handle_chat_retry` to trim history and re-invoke `chat_respond` (chat_handler.py:handle_chat_retry())
- [feat] Modify `create_chat_tab` to accept `handle_chat_retry_fn` parameter (ui_components.py:create_chat_tab():14)
- [feat] Call `chatbot_display.retry()` with `handle_chat_retry_fn` when provided (ui_components.py:create_chat_tab():83)
- [feat] Import `handle_chat_retry` from `chat_handler` (app.py:5)
- [feat] Pass `handle_chat_retry` to `create_chat_tab` in `create_app` (app.py:create_app():29)

app.py

@@ -4,7 +4,7 @@ A comprehensive AI platform with chat and image generation capabilities.
 """
 
 import gradio as gr
-from chat_handler import handle_chat_submit
+from chat_handler import handle_chat_submit, handle_chat_retry
 from image_handler import handle_image_generation
 from ui_components import (
     create_main_header,
@@ -27,7 +27,7 @@ def create_app():
         with gr.Tabs() as tabs:
             
             # Chat tab
-            create_chat_tab(handle_chat_submit)
+            create_chat_tab(handle_chat_submit, handle_chat_retry)
             
             # Image generation tab  
             create_image_tab(handle_image_generation)

chat_handler.py

@@ -193,3 +193,65 @@ def handle_chat_submit(message, history, system_msg, model_name, max_tokens, tem
         # Update history with the current partial response and yield it
         current_history = history + [{"role": "assistant", "content": assistant_response}]
         yield current_history, ""
+
+
+def handle_chat_retry(history, system_msg, model_name, max_tokens, temperature, top_p, retry_data=None):
+    """
+    Retry the assistant response for the selected message.
+    Works with gr.Chatbot.retry() which provides retry_data.index for the message.
+    """
+    # Guard: empty history
+    if not history:
+        yield history
+        return
+
+    # Determine which assistant message index to retry
+    retry_index = None
+    try:
+        retry_index = getattr(retry_data, "index", None)
+    except Exception:
+        retry_index = None
+
+    if retry_index is None:
+        # Fallback to last assistant message
+        retry_index = len(history) - 1
+
+    # Trim history up to the message being retried (exclude that assistant msg)
+    trimmed_history = list(history[:retry_index])
+
+    # Find the most recent user message before retry_index
+    last_user_idx = None
+    for idx in range(retry_index - 1, -1, -1):
+        if trimmed_history[idx].get("role") == "user":
+            last_user_idx = idx
+            break
+
+    # Nothing to retry if no prior user message
+    if last_user_idx is None:
+        yield history
+        return
+
+    # Message to retry and prior conversation context (before that user msg)
+    message = trimmed_history[last_user_idx].get("content", "")
+    prior_history = trimmed_history[:last_user_idx]
+
+    if not message.strip():
+        yield history
+        return
+
+    # Stream a new assistant response
+    response_generator = chat_respond(
+        message,
+        prior_history,
+        system_msg,
+        model_name,
+        max_tokens,
+        temperature,
+        top_p
+    )
+
+    assistant_response = ""
+    for partial_response in response_generator:
+        assistant_response = partial_response
+        current_history = trimmed_history + [{"role": "assistant", "content": assistant_response}]
+        yield current_history

ui_components.py

@@ -11,7 +11,7 @@ from utils import (
 )
 
 
-def create_chat_tab(handle_chat_submit_fn):
+def create_chat_tab(handle_chat_submit_fn, handle_chat_retry_fn=None):
     """
     Create the chat tab interface.
     """
@@ -81,6 +81,15 @@ def create_chat_tab(handle_chat_submit_fn):
             outputs=[chatbot_display, chat_input]
         )
 
+        # Enable retry icon and bind handler if provided
+        if handle_chat_retry_fn is not None:
+            chatbot_display.retry(
+                fn=handle_chat_retry_fn,
+                inputs=[chatbot_display, chat_system_message, chat_model_name, 
+                        chat_max_tokens, chat_temperature, chat_top_p],
+                outputs=chatbot_display
+            )
+
 
 def create_chat_tips():
     """Create the tips section for the chat tab."""
@@ -289,3 +298,5 @@ def create_footer():
     
     **Built with ❤️ using [HF-Inferoxy](https://nazdridoy.github.io/hf-inferoxy/) for intelligent token management**
     """)
+
+