Implement exception throttling for jobs

[JaZo]

Detailed description

We should implement the exception throttling middleware on the jobs.

Context

Our Elasticsearch instance experienced some downtime and all jobs kept trying, causing them all to fail and our logging to be flooded. When we enable the middleware, they are throttled and wait for the instance to come back up.

[JaZo]

Version 0.6.0 added support for specifying job classes, so you can create your own jobs with the middleware you'd like.