NameNode replicas not configurable

[lbornov2]

When we deploy the hbase helm chart, the namenode replicas should be configurable. Currently, it is currently hard-coded to 1.

https://github.com/Gradiant/bigdata-charts/blob/master/charts/hdfs/templates/hdfs-nn-statefulset.yaml

This is a single point of failure. See output of kubectl get pods | grep hbase:

[ec2-user@ip-xyz ~]$ kubectl  get pods | grep hbase
dev-hbase-hbase-master-0                             1/1     Running   1          16m
dev-hbase-hbase-master-1                             1/1     Running   0          14m
dev-hbase-hbase-master-2                             1/1     Running   0          14m
dev-hbase-hbase-regionserver-0                       1/1     Running   2          16m
dev-hbase-hbase-regionserver-1                       1/1     Running   0          14m
dev-hbase-hbase-regionserver-2                       1/1     Running   0          14m
dev-hbase-hdfs-datanode-0                            1/1     Running   0          16m
dev-hbase-hdfs-datanode-1                            1/1     Running   0          15m
dev-hbase-hdfs-datanode-2                            1/1     Running   0          15m
dev-hbase-hdfs-httpfs-5fff67bcb8-8mkpj               1/1     Running   0          16m
dev-hbase-hdfs-namenode-0                            2/2     Running   2          16m
dev-hbase-zookeeper-0                                1/1     Running   0          16m
dev-hbase-zookeeper-1                                1/1     Running   0          16m
dev-hbase-zookeeper-2                                1/1     Running   0          16m

We are deploy this chart:

https://artifacthub.io/packages/helm/gradiant/hbase