Improve the etcd liveness probe in helm chart

[bsctl]

In etcd 3.5, the /health endpoint returns false if there is no etcd leader or leader-election is currently going on. If there is no quorum, etcd will get restart by kubelet but etcd's restart isn't a right behaviour.

Starting from etcd 3.5.3+ support for checking the health of a specific member (as opposed to the cluster as a whole) has been introduced. Upstream kubeadm is switching to using this for static pod health checks, as there's no point in restarting the pod if the cluster as a whole is unhealthy - as a matter of fact it may actually make it worse. We should do the same for Kamaji.

References:

kubernetes/kubernetes#110072
etcd-io/etcd#13399

...
      containers:
        - name: etcd
          image: quay.io/coreos/etcd:v3.5.4
          livenessProbe:
            failureThreshold: 8
            httpGet:
              path: /health?serializable=true
              port: 2381
              scheme: HTTP
            initialDelaySeconds: 10
            periodSeconds: 10
            timeoutSeconds: 15
          startupProbe:
            failureThreshold: 24
            httpGet:
              path: /health?serializable=true
              port: 2381
              scheme: HTTP
            initialDelaySeconds: 10
            periodSeconds: 10
            timeoutSeconds: 15