This updates RubocopTool to subclass the newer BaseTool, and to use

the newer features it offers (dependencies, improved file extension

handling, centralized base command building, and new comment options).
The parsing has been redone to use the Rubocop JSON output, which

constains details such as line/column ranges and error codes. It also

tells Rubocode to provide URLs to any relevant docs when reporting

errors, and forces it not to comment on filenames (which can fail

depending on the temp filename that's created).
If we're unable to parse the JSON payload, we split the lines and use

the first line as an error. This seems to be the most reliable option,

as subsequent lines can end up with stack traces or "successful" JSON

payloads.

Unit tests pass on Python 2.7 and 3.x.