ceph wait callbacks

[jrse]

Currently the plugin is using the librados call:
wait_for_complete_and_cb =>
Block until an operation completes and callback completes

This means it is in memory on all replicas and can be read.

There is another wait for complete method with a higher degree of safety, but slower E.g.

rados_aio_wait_for_safe_and_cb
=> Block until an operation is safe and callback has completed

This means it is on stable storage on all replicas.

Should this be configurable?