Add assertion helpers for prepared requests

[DevL]

The assertions function/functions are invoked with a PreparedRequest object that can be unwieldy to work with. For example, a URL containing encoded query parameters aren't fun to work with.

Python's standard library urllib.parse to the rescue!
We can use urllib.parse.urlparse combined with urllib.parse.parse_qs to break the URL apart into its components and then provide nicer abstractions to work with when writing assertions functions.

>>> from urllib.parse import parse_qs, urlparse
>>> url = "https://www.example.com/some/endpoint?name=Lennart&beverages=tea%2Ctonic%2Cbeer"
>>> urlparse(url)
ParseResult(scheme='https', netloc='www.example.com', path='/some/endpoint', params='', query='name=Lennart&beverages=tea%2Ctonic%2Cbeer', fragment='')
>>> parsed = urlparse(url)
>>> parsed.query
'name=Lennart&beverages=tea%2Ctonic%2Cbeer'
>>> parse_qs(parsed.query)
{'name': ['Lennart'], 'beverages': ['tea,tonic,beer']}

[DevL]

parse_qs returns a dict where the keys are lists as the same query parameter can be occur multiple times in the same string. This needs to be handled. At the same time, it'd be helpful to unpack single values so that assertions doesn't have to wrap the expected value in a list.

Params with missing values should be removed.

>>> from urllib.parse import parse_qs
>>> query_string = 'a=1&a=2&b=3&c='
>>> parse_qs(query_string)
{'a': ['1', '2'], 'b': ['3']}

The above should be accessible like

parsed_request = ParsedRequest(prepared_request)
assert parsed_request.query == {'a': ['1', '2'], 'b': '3'}