Support for a list of urls for web-extract

[kevinschaper]

You can definitely take or leave this feature request. For the MIC ingest, I think the ideal workflow would be to provide a list of urls to web-extract and receive a single pair of kgx node & edge tsv files, and keep the url of the pages in those files.

That might be a whole bundle of way too specific feature requests though, so the alternative would probably be just patching in file -o support for kgx tsv. (I saw a TODO in there) and then they can be merged, and then we can handle iterating on the urls and merging the output.