post-install job uses hard-coded cluster.local DNS suffix

[kaiburjack]

Describe the bug
When using a Kubernetes setup with a custom DNS cluster domain suffix (like is required with a GKE VPC-scoped Cloud DNS), the post-install job will fail because of a hard-coded cluster.local domain part.

While it is possible to override the DNS clusterDomainSuffix as a Helm value dns.clusterDomainSuffix, this value apparently is not used by the post-install job (only in the ConfigMap used by the couchdb pods).

Version of Helm and Kubernetes:
(irrelevant)

What happened:
(see description above)

What you expected to happen:
Helm chart working with a custom cluster domain suffix.

How to reproduce it (as minimally and precisely as possible):
Install couchdb in a Kubernetes cluster with a non-default DNS suffix (i.e. not cluster.local).

Anything else we need to know: