Add support for parsing DOCX files

[hawkeyexl]

Lots of writers author in Word format. Internally, Word files use an XML document representation that, with relevant metadata, can be parsed. Because the underlying content is much more structured than something like markdown, we'll have to approach the style config and parsing differently, but we should still be able to identify heading levels, bolding, and more.

[hawkeyexl]

1. Convert DOCX to HTML: https://github.com/mwilliamson/mammoth.js
2. Convert HTML to Markdown: https://github.com/mixmark-io/turndown
3. Parse Markdown with default style options.