Fix: Enable Gradio queue for generator functions and suppress T5 warnings

Fixes:
✅ Added demo.queue() to enable generator function support
✅ Added warnings filter to suppress T5 tokenizer FutureWarning
✅ Maintains 100% free functionality

The generator functions (yield statements) require queue to be enabled in Gradio.
This allows real-time progress updates to the user during processing.

🤖 Generated with Claude Code
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

app.py

@@ -1,10 +1,14 @@
 import os
 import re
+import warnings
 import gradio as gr
 import fitz
 from transformers import pipeline
 import torch
 
+# Suppress T5 tokenizer warnings
+warnings.filterwarnings("ignore", category=FutureWarning, module="transformers")
+
 # Initialize models
 print("Loading AI models...")
 device = 0 if torch.cuda.is_available() else -1
@@ -581,4 +585,5 @@ with gr.Blocks(title="Ultimate Exam Prep - Study Guide Generator", theme=gr.them
     """)
 
 if __name__ == "__main__":
+    demo.queue()  # Enable queue for generator functions
     demo.launch()