I am trying to find an updated way to convert docx and pptx files to PDF in Lambda in a long-term, sustainable manner. I am creating a data ingestion pipeline and Textract cannot work with docx or pptx.