Optional SIGTERM at Job Failure

I would like to THANK YOU for this brilliant piece of software!

I have a use case in which `supercronic` acts as the entry-point of a container. That leads to the necessity to restart the container if the job that is run by `supercronic` fails.

The issue does not name any files, tests, or entry points. Clarify when supercronic should send SIGTERM after a job failure, how this should be configured, and what the expected container behavior and tests should demonstrate.

Written by the indexing model from the issue text.