Node Drain Request/Status API

[andrewd-zededa]

The base of an upcoming PR to enable deferring node reboot and shutdown operations. In an Edge-Node-Cluster each node can hold a replica copy of a cluster volume. If the other nodes 1 and 2 in a cluster have recently recovered from a failure they could be rebuilding their replicas (1.x, 1.y, 2.x, 2.y) off of the only remaining readable copies (3.x, 3.y) on node 3. This API is to enable zedagent,nodeagent,baseosmgr to request a drain of replicas off of an edge-node (rebuild completion) before allowing a node outage.

[rene]

@andrewd-zededa , you have some Yetus errors to fix.

[andrewd-zededa]

Forgot to actually commit the kube-status change in nodedrain.go, its there now.

[andrewd-zededa]

@eriknordmark changes pushed up to address review questions. Please let me know if any further changes are needed.