feat(docx text extraction): extract all the text from the uploaded docx file

fix(pdfplumber): replace the pdfplumber package and implementation with pymupdf

feat(pdf text extraction): extract all the text from the uploaded pdf file

updated requirements.txt to only have openai; get rid of old imoprt statements that are not needed (i.e. spaces, transformers, torch, etc.)

Add openai package to requirements.txt

update requirements.txt to not have versions

add accelerate pip package and add device_map='auto'

add transformers pip package and remove TextIteratorStreamer from imports

add torch