[chart/redis-ha][BUG] client writes errors on redis rolling update (failover not triggered)

[slambour]

Describe the bug
During redis-ha rolling updates, Redis clients occasionally experience connection errors (~10s interruption) when a master pod is terminated.

To Reproduce
Steps to reproduce the behavior:

1. Deploy redis with 3 replicas + 3 sentinels
2. Deploy a redis client with a 1 write / second
3. Delete the pod master (or a statefulset restart)
4. See client write errors during 10 seconds

Expected behavior
I expect to see a master change on the client without write errors

Additional context
I have tested a fix on a cluster here is the commit on my fork slambour@4ef7e3b