PDF extraction improvements

[davidmezzetti]

Modify the PDF file extraction process as follows:

• Use dateutil to parse and format the published/date field
• Extract content from tables
• Build uid off the title but fallback to doi if title is not found
• Add tag of PDF to articles