Pre-processing for vectorDB

1. Function:

• Before converting to vectorDB, extract TXT from various file formats: PDF, Excel, PPT, Word, PNG, JPG.
• When converting to TXT, display the file title and saved path.

2. OCR:

paddleocr-gpu / cuda12.6

3. Functions:

1. main.py
2. ppt2pdf.py : convert *.ppt / *.pptx to pdf format by libreoffice