Pytube

* Fixes RegexMatchError in n cipher
* Added documentation for some new functionality
* `last_updated` will now return the raw text if it can't parse a datetime (e.g. "X days ago")
* change default innertube client for `Search` objects to `WEB`

* User InnerTube in place of `get_video_info` url

* Added some additional base parameters for innertube requests.

* Added Oauth support for innertube client

* Add exception for age-restricted videos which can no longer be accessed without using auth.

* Carved out and simplified code where possible due to API changes.

* Added renderer catch -- fixes 1068

* Additional channel name support for URL-encoded names.

* Updated test mocks, removed region-locked test because that functionality no longer works.

Due to significant changes in how video info is accessed, certain YouTube object attributes no longer exist, which could cause breaking changes in code bases that rely on those attributes, hence the major version release.

* Add catch for suggested search results; accounts for edge case of no views on result.

* Added exception handling for incorrect cached js files.

* Now allows you to **actually** set filenames, instead of doing partial overrides to create safe filenames.

* Innertube improvements, and skeleton code for future innertube work

* Update mock so that stream raises HTTPError, rather than head, and simplified some other mocks.

* pep8-naming updated, adding a new ignore

* Removed special character from __author__ attribute.

* Changed -v CLI arg to have a single setting, rather than multiple.

* Add retry functionality for IncompleteRead errors.

* Extract contentLength from info where possible.

* Mock open in final streams test to prevent file from being written.

* Exception handling for accessing titles of private videos.

* Emulates the js player to calulate the value of n to prevent download slowdowns

* Adds some additional debugging to HTMLParseErrors

* Adds a helper for generating new html json files

* Parser improvement