Add a method for scraping specific bills and events

[hancush]

It's currently impossible to scrape a single bill without also scraping lots of other bills. Ditto for events. Let's look into a way to update specific bills and events by passing in matter/event IDs, respectively.

[hancush]

Preliminary design

Add a fetch method to the API scraper that accepts the appropriate ID of the bill/event to retrieve.

legistar.bills

class LegistarAPIBillScraper(LegistarAPIScraper) :
    ...
    fetch = partialmethod(endpoint, '/matters/{}')

In the downstream scrapers, break the matter formatting and related information that currently takes place in the scrape method into a separate method or methods (i.e., for bill formatting and action formatting).

Add a keyword argument that defaults to None but can be set to an ID, should the user desire to scrape specific matters/events.

chicago.bills, for example

class ChicagoBillScraper(LegistarAPIBillScraper):
    ...
    def scrape(self, window=3, matter_id=None):
        if matter_id:
            matter = self.fetch(matter_id)
            yield self.format_matter(matter)

        else:
            n_days_ago = datetime.datetime.utcnow() - datetime.timedelta(float(window))
            for matter in self.matters(n_days_ago):
                yield self.format_matter(matter)

[hancush]

Done for bills in #60.