Removing ceph-fs unit doesn't cleanup from cluster

[Ponnuvel Palaniyappan]

After removing a ceph-fs unit with `juju remove-application`, it's still not cleaned up from Ceph:
```
  cluster:
    id: 447eda2c-ca9e-11ee-b47d-b7246819e2f0
    health: HEALTH_WARN
            insufficient standby MDS daemons available

  services:
    mon: 1 daemons, quorum juju-1c1b87-reef-0 (age 2d)
    mgr: juju-1c1b87-reef-0(active, since 2d)
    mds: 1/1 daemons up
    osd: 4 osds: 3 up (since 47h), 3 in (since 46h)

  data:
    volumes: 1/1 healthy
    pools: 19 pools, 159 pgs
    objects: 245 objects, 459 KiB
    usage: 162 MiB used, 30 GiB / 30 GiB avail
    pgs: 159 active+clean
```

[Luciano Lo Giudice (lmlogiudice)]

Indeed, there doesn't seem to be any hook implementation for unit departure.