Better parser framework

[Highfire1]

wants:

• much better caching that can be easily defined
• better long term storage solution than storing html blobs in a sqlite db
• be more extensible to support easily adding other parsers (such as #1 )

[Highfire1]

also selenium and pymupdf are slow and bad