derive list of external URLs

[novoid]

Add an additional command line parameter similar to --external-urls $FILE. This $FILE gets populated with CSV content like:

http://foo.example.com/an/old/path; id:2020-04-12-article-id
https://bar.example.com; id:another-id

This can be used as input for any external URL checker in order to find broken links.

Implementation might relate to #41 and #44 which also might potentially extract data without generating the actual blog data (not decided yet).

[novoid]

• [x] decide: accept duplicated entries for combinations and/or URLs only?
• [x] decide: only as a by-product of a normal lazyblorg generation run or also as a stand-alone feature?

[novoid]

An approach could be to combine this with #44 in the following way:

A command line option results in multiple files generated in parallel to a normal blog generation run:

1. ${BLOG_NAME}_URLs.csv: a CSV file with tabs as separators (which don't appear in the data) with lines (one per URL) with:

• ID of the article
• the optional description used for the description (sanitized for tabs)
• an external URL referred in the article (sanitized for tabs)

2. ${BLOG_NAME}_files.csv: a CSV file with tabs as separators (which don't appear in the data) with lines (one per referred file) with:

• ID of the article
• the optional description used for the description (sanitized for tabs)
• an external file reference in the article (sanitized for tabs): mostly image files
  • [ ] absolute file name versus relative or ts-basename?
  • [ ] how to deal with non-matching basenames within the Org file which got fixed by the built-in "fuzzy" search?