Add Manila CephNFS support

[fmount]

This patch introduces a set of basic tasks to perform the adoption of Manila with a CephNFS backend.
Before performing the actual adoption, a new CephNFS cephadm based cluster must be created.
For this reason, the manila_nfs tasks require a few parameters as input:

1. A new CephNFS VIP where the CephIngress daemon (made by Haproxy and keepalived) is created;
2. A set of Ceph target_nodes where the "nfs" label is added: these nodes are supposed to host the new CephNFS cluster and must be different from controller nodes that are going to be decommisioned;
   3. The TripleO managed Ganesha VIP: this input can be actually retrieved by the existing TripleO deployment, but we can gather this input as part of the next iteration on this patch

After the new CephNFS cluster is created, it is possible to build a manila-share with the expected NFS configuration.

[fmount]

@lkuchlan I might need your help to testproject this use case and define the set of parameters we need to pass as overrides when NFS is set as a backend.
@gouthampacha I start doing some sanity check against this code and try to reduce (as much as possible) the number of overrides we need to setup in the job.
For the reason above, I'm holding this change for now.

[softwarefactory-project-zuul]

Build failed (check pipeline). Post recheck (without leading slash)
to rerun all jobs. Make sure the failure cause has been resolved before
you rerun jobs.

https://softwarefactory-project.io/zuul/t/rdoproject.org/buildset/c17fe76e4a084abc8ee5afb3fe05352e

❌ adoption-standalone-to-crc-ceph FAILURE in 2h 37m 42s
❌ adoption-standalone-to-crc-no-ceph FAILURE in 2h 41m 37s

[fmount]

@gouthampacha @lkuchlan I think we're still missing something here. As a prereq, when the overcloud still exists, we need to propagate the StorageNFS network to the Ceph target nodes. For this reason we might need an additional change where we do that beforehand.
I'm thinking to create an additional PR where we can discuss and focus on the following:

1. build the appropriate baremetal file from the undercloud
2. add the StorageNFS network to the Ceph target nodes
3. run overcloud node provision --network-config
4. verify the network exists in the target nodes

While the target nodes should be passed as input, the above would also remove the requirement of getting the new VIP.

[softwarefactory-project-zuul]

Build failed (check pipeline). Post recheck (without leading slash)
to rerun all jobs. Make sure the failure cause has been resolved before
you rerun jobs.

https://softwarefactory-project.io/zuul/t/rdoproject.org/buildset/bb8e08db65c045ac98ed9d23bdc24d15

✔️ adoption-standalone-to-crc-ceph SUCCESS in 2h 55m 38s
❌ adoption-standalone-to-crc-no-ceph RETRY_LIMIT in 26m 29s