backend - handle openai errors gracefully

[yevkim]

Openai might return content moderation related query if user asks something explicit

explore this code
try:
# Call your LLM chain
for chunk in chain_with_history.stream({"query": input_text}, config=config):
if hasattr(chunk, 'content'):
yield chunk.content
full_response.append(chunk.content)
except ValueError as e:
# Check if the error message indicates content filtering
if "content filter being triggered" in str(e).lower():
yield "I’m sorry, but I cannot provide an answer to that request."
return
else:
raise