Consider adding ability to wait for all volumes to be detached from node before rebooting by agent

[invidian]

With storage.k8s.io/v1.VolumeAttachment resource, we could now quite effectively resolve coreos/container-linux-update-operator#192 I think. Given that this feature is driven upstream, I guess it would make sense to support it in tree.

The rough implementation would be:

• drain
• wait for all volumes to be detached from a given node (opt in initially, with optional force after deadline)
• reboot

Old issue: coreos/container-linux-update-operator#192

Current workaround PR: #62

[invidian]

Just realized we have #30 already, so we can close this. Sorry for the noise.