Fix: Gradio UI image display for text-to-LEGO generation

Problem: Rendered LEGO images weren't showing after successful rendering
Root cause: Return value mismatch between unified_input_handler() and Gradio outputs

Changes:
- unified_input_handler() now returns 6 values (was 5)
- Added LDR text output as 5th return value
- Added rendered_image as 6th return value
- Updated Gradio output mapping to include render_result (Image component)

Affected functions:
- unified_input_handler(): Returns (renderings, part_list, status, ldr_data, ldr_text, rendered_image)
- process_ldr_file(): Returns (..., None, None) for file upload flow
- process_ldr_from_path(): Returns (..., None, None) for path-based flow

UI Flow:
1. User enters "red car" → CLIP retrieval → GPT generation → Blender render
2. Image now displays correctly in "Rendering Result" section
3. LDR text displays in "Generated LDR Content" textbox

This completes the text-to-LEGO visualization pipeline.

code/demo.py

@@ -152,10 +152,10 @@ def process_ldr_file(file, process_for_model=True):
                           If False, skip numerical conversion (only extract parts for visualization).
 
     Returns:
-        Tuple of (renderings, part_list, status, process_ldr_data, None)
+        Tuple of (renderings, part_list, status, process_ldr_data, None, None)
     """
     if not file:
-        return None, None, "Please upload an LDR file", None, None
+        return None, None, "Please upload an LDR file", None, None, None
 
     # Read LDR content
     with open(file.name, 'r') as f:
@@ -202,7 +202,7 @@ def process_ldr_file(file, process_for_model=True):
         print(f"🔍 [DEBUG] Skipping numerical conversion (process_for_model=False)")
 
     print(f"🔍 [DEBUG] Final process_ldr_data: {'None' if process_ldr_data is None else 'has data'}")
-    return renderings, part_list, f"File loaded, {len(part_names)} valid parts identified", process_ldr_data, None
+    return renderings, part_list, f"File loaded, {len(part_names)} valid parts identified", process_ldr_data, None, None
 
     # except Exception as e:
     #     return None, None, f"File processing failed: {str(e)}", None, None
@@ -218,10 +218,10 @@ def process_ldr_from_path(ldr_path, process_for_model=False):
                           If False (default), skip numerical conversion for visualization-only.
 
     Returns:
-        Tuple of (renderings, part_list, status, process_ldr_data, None)
+        Tuple of (renderings, part_list, status, process_ldr_data, None, None)
     """
     if not os.path.exists(ldr_path):
-        return None, None, f"LDR file not found: {ldr_path}", None, None
+        return None, None, f"LDR file not found: {ldr_path}", None, None, None
 
     # Create a mock file object to reuse process_ldr_file logic
     class MockFile:
@@ -249,7 +249,7 @@ def unified_input_handler(file, text_query):
     # Case 2: Text query (neural generation)
     elif text_query and text_query.strip():
         if not CLIP_AVAILABLE:
-            return None, None, "❌ Text-to-LEGO feature is not available (generation module not loaded)", None, None
+            return None, None, "❌ Text-to-LEGO feature is not available (generation module not loaded)", None, None, None
 
         try:
             # Generate LDR design from text
@@ -261,7 +261,7 @@ def unified_input_handler(file, text_query):
             result = retriever.get_best_match(query)
 
             if result is None or not result.get("ldr_exists", True):
-                return None, None, f"❌ Could not generate design for '{query}'", None, None
+                return None, None, f"❌ Could not generate design for '{query}'", None, None, None
 
             ldr_path = result["ldr_path"]
             confidence = result["similarity"]
@@ -277,7 +277,7 @@ def unified_input_handler(file, text_query):
 
             # Check if numerical conversion succeeded
             if process_ldr_data is None:
-                return None, None, f"❌ Failed to convert LDR to model format (missing label mappings)", None, None
+                return None, None, f"❌ Failed to convert LDR to model format (missing label mappings)", None, None, None
 
             # Generate new LDR using GPT model (GPU-accelerated)
             new_ldr_filename = f"generated_{uuid.uuid4()}.ldr"
@@ -295,17 +295,21 @@ def unified_input_handler(file, text_query):
             # Update status message with generation info
             enhanced_status = f"✨ Generated from car_{car_id} (confidence: {confidence*100:.1f}%)\n🤖 GPT model created new assembly sequence\n{status}"
 
-            return renderings, part_list, enhanced_status, process_ldr_data, rendered_image
+            # Read generated LDR content for display
+            with open(new_ldr_path, 'r', encoding='utf-8') as f:
+                ldr_text = f.read()
+
+            return renderings, part_list, enhanced_status, process_ldr_data, ldr_text, rendered_image
 
         except Exception as e:
             import traceback
             error_msg = f"❌ Design generation failed: {str(e)}\n{traceback.format_exc()}"
             print(error_msg)
-            return None, None, error_msg, None, None
+            return None, None, error_msg, None, None, None
 
     # Case 3: No input
     else:
-        return None, None, "⚠️ Please upload an LDR file OR enter a text description", None, None
+        return None, None, "⚠️ Please upload an LDR file OR enter a text description", None, None, None
 
 
 import traceback  # 导入traceback，用于打印完整堆栈
@@ -474,7 +478,7 @@ with gr.Blocks(
     upload_btn.click(
         fn=unified_input_handler,
         inputs=[ldr_file, text_input],
-        outputs=[part_renderings, part_list, status_msg, original_ldr, predicted_ldr]
+        outputs=[part_renderings, part_list, status_msg, original_ldr, predicted_ldr, render_result]
     )
 
     predict_btn.click(