For exceptions that happen on WorkRunner threads, we log an error rather than let them be forwarded to the threads' UncaughtExceptionHandler. This PR adds the ability to configure
that behaviour, so that it's possible to do something else - like crash the application, for instance.
I feel like this is a workable approach, but I'm not sure that it's any better than encouraging users to set the UncaughtExceptionHandler for the threads in the WorkRunners.
For exceptions that happen on WorkRunner threads, we log an error rather than let them be forwarded to the threads' UncaughtExceptionHandler. This PR adds the ability to configure that behaviour, so that it's possible to do something else - like crash the application, for instance.