Support Markdown in URL matching

[pmonks]

When URL matching retrieves the content of a URL in order to perform full text license matching, it would be ideal if text/markdown responses were supported (i.e. by converting the Markdown to text/plain, and then performing license text matching on that).

[pmonks]

The commonmark-java library looks like a good way to achieve this, though it will be a new dependency.