Switch from PyPDF to pdfminer

[Eurybiadan]

PyPDF appears to not support new pdf versions (>=1.4). Switching to pikepdf should provide better support as it appears better maintained.

[Eurybiadan]

Scratch that, pikepdf can't do text extraction. Perhaps https://github.com/pdfminer/pdfminer.six

[iraklis10]

Script run fine after running pip3 install pdfminer.six on a Windows 10 machine command line