Support extraction without OCR and small data volume

Changes

• Ability to extracting text through pdftotext (without OCR)
• Improved interface for using pdf2dataset as a lib and not through CLI
• Reduced disk IO, document read only once
• Add specific mode for small volumes of documents

Fix some warnings

Changes

• Small improvements (remove some deprecated methods and text of one warning)
• Add some important tests
• Improve README

Initial Version

Features

• Conversion of a whole subdirectory with PDFs documents into a pandas DataFrame
• Support of parallel and distributed computing through ray
• Incremental writing of resulting DataFrame, to save memory
• Ability to keep processing progress and resume from it
• Error tracking of faulty documents
• Use OCR for extracting text through pytesseract and pdf2image
• Initial, easy to use, CLI
• Custom behaviour through parameters (number of CPUs, text language, etc)