Allow and perform storageBackend Migrations from couchstore to Magma

Base Information: https://review.couchbase.org/c/ns_server/+/192124

During Bucket reconciliation, we should allow the modification of a bucket between Couchstore and Magma (in either direction), only for Couchbase Server 7.6

During Pod reconciliation, if the cluster version is greater than 7.6, a check should be made to each bucket to ensure that the storageBackend of each node found at

curl -u Administrator:asdasd

http://127.0.0.1:9000/pools/default/buckets/foo | jq

'.nodes[].storageBackend' "couchstore" "couchstore"

matches the storage backend for the bucket found at

curl -u Administrator:asdasd

http://127.0.0.1:9000/pools/default/buckets/foo | jq '.storageBackend'

"magma"

If these values do not match, the pod should be swap rebalanced out of the cluster and replaced with a new pod.

This step should occur after Pod changes but before API changes in the Reconciliation Loop.  (We don't want to double work if pod changes result in a swap rebalance anyway)

This should check all buckets in the system, not just K8S managed buckets. You can use

GET /pools/default/buckets

to retrieve the list of buckets to check.