Support capture of Java heap dump

[planetf1]

If a JVM dies due to running out of memory when running in a k8s environment the jvm will exit, the process will exit, and the container will go away.

Even if appropriate flag is set to capture heap dumps, it will only get written to ephemeral storage and will not be retrievable to allow debugging of the problem

We could arrange for the heap dump to be written to a persistent volume which could be read after the failure.

This would be a useful debugging aid to add to the charts, and a 'best practice' recommendation for those deploying to k8s using their own scripts/charts.

Whilst this issue may not be a heap issue, it's caused a discussion on how we might debug:
See #235