Fix processing indicator and improve system prompt

- Fix processing indicator appearing in old messages - now shows only once at bottom
- Remove custom processing function, use simple st.spinner
- Simplify system prompt with general coding practices instead of specific error fixes
- Focus on fundamental pandas/Python best practices for better LLM code generation

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

app.py

@@ -759,9 +759,6 @@ def show_custom_response(response):
             
         return {"is_image": False}
 
-def show_processing_indicator(model_name, question):
-    """Simple processing indicator"""
-    st.info(f"Processing with {model_name}...")
 
 # Chat history
 # Display chat history
@@ -806,12 +803,10 @@ for response_id, response in enumerate(st.session_state.responses):
                 st.success("Thanks for your feedback!")
                 st.rerun()
 
-    # Show processing indicator if processing
-    if st.session_state.get("processing"):
-        show_processing_indicator(
-            st.session_state.get("current_model", "Unknown"), 
-            st.session_state.get("current_question", "Processing...")
-        )
+# Show processing indicator if processing (only once at the bottom)
+if st.session_state.get("processing"):
+    with st.spinner(f"Processing with {st.session_state.get('current_model', 'Unknown')}..."):
+        pass
 
 # Chat input with better guidance
 prompt = st.chat_input("💬 Ask about air quality trends, compare cities, or request visualizations...", key="main_chat")

src.py

@@ -310,14 +310,13 @@ SAFETY & ROBUSTNESS RULES:
 - Validate date ranges exist in data
 - Use proper string formatting for answers with units (μg/m³)
 
-CRITICAL: PANDAS SYNTAX FIXES:
-- ALWAYS convert pandas/numpy values to int before using as list indices
-- Example: calendar.month_name[int(month_value)] NOT calendar.month_name[month_value]  
-- Use int() conversion for ANY value used as index: int(row['month']), int(max_idx), etc.
-- When accessing pandas iloc results, wrap in int(): int(df.loc[idx, 'column'])
-- CORRECT groupby syntax: df.groupby([df['col1'], df['col2'].dt.year]) NOT df.groupby(['col1', 'col2'].dt.year)
-- Always reference DataFrame when accessing columns: df['column'].dt.year NOT 'column'].dt.year
-- Use proper DataFrame column references in all operations
+CRITICAL CODING PRACTICES:
+- Always convert pandas/numpy objects to proper Python types before operations
+- Use descriptive variable names (avoid single letters in complex logic)
+- Ensure all variables are defined before use
+- Convert datetime/period objects to appropriate types (.astype(str), .astype(int)) 
+- Reference DataFrame properly in all column operations: df['column'] not 'column'
+- Use proper data type checking and conversion for all operations
 
 TECHNICAL REQUIREMENTS:
 - Save final result in variable called 'answer'