Improve fuzzy url matching throughout application

[u8sand]

Might make sense to make a dedicated module and tests for this. Some ideal features of good URL matching:

• [ ] url in search
• [ ] search in url
• [ ] try chopping out common differentiating parts of a url

  ^(?P<proto>.+?)://(?P<url>.+?)(\?(?P<server_vars>.+?))?(#(?P<client_vars>.+?))?$

[u8sand]

In general, searches could be more efficient for things like this and more if we migrate our database to PostgreSQL and take advantage of its features. (https://docs.djangoproject.com/en/2.1/ref/contrib/postgres/search/#searchrank)

Another thing which could be done to help the search along is to explicitly split urls into their components:

import urllib
parsed_url = urllib.parse.urlparse(url)

urllib.parse.urlparse('http://me:mine@google.com/hello/world?q=hi&b=bye#c=f')
# ParseResult(scheme='http', netloc='me:mine@google.com', path='/hello/world', params='', query='q=hi&b=bye', fragment='c=f')