Pass the not RAG queries directly to LLM to answer!

[amindadgar]

In case of enable_answer_skipping equal to False

This feature was there before but as we removed the crewai tool calling and replaced it with langchain, it's removed.