Intelligent parsing of various documents including PDF, Word, PPT, etc., applicable for machine learning, large model corpus production, RAG and other scenarios