Ignore job failures

[davetron5000]

Problem

When we cannot connect to Redis, which is frequent with RedisCloud, the monitoring jobs fail. Since they are run periodically, there's no sense retrying them, so we end up with a lot of grinding.

Solution

Add option to log and ignore them.

Addendum

MonitorJob had no test coverage, and adding it was tricky because it relies on a lot of global state. Thus, I sadly had to introduce a mocking library in order to get test coverage. With that coverage, I refactored the class to not create every single checker and notifier every single time, instead only creating what is needed. This change isn't strictly needed for this, but it allowed the tests to be a lot easier to deal with.

[davetron5000]

For whatever reason, the Mocha setup instructions differ from the code and neither works enough to set up Mocha in the tests properly, so it seems I have to do all this manually. Which is fine—I'd like to keep mocking at a minimum