ft: ZENKO-781 scale out backbeat services

[jonathan-gramain]

• Modify number of kafka partitions created to nodeCount

• Set all scalable backbeat components replica count to nodeCount

[giacomoguiulfo]

Since most of the backbeat deployments share this value, maybe it should be specified in the values.yaml file. Then it could be templated with something like:

        	livenessProbe:
{{ toYaml .Values.livenessProbe | indent 12 }}
        	readinessProbe:
{{ toYaml .Values.readinessProbe | indent 12 }}

[jonathan-gramain]

For now I keep those changes as an experiment to see how it goes on the CI, but if it helps I'll do this change!

[ssalaues]

I think the periodDelay of something like 10-15 would be very useful in this PR as well since the dependency services (kafka and zookeeper) usually take the longest to come up. This way the exponential CrashLoopBackoff doesn't cause the pods to quickly build up a long retry period.

[jonathan-gramain]

Since the problem in this PR does not seem related to this initial delay but to conflicts with other pending PRs, let's keep those useful hints aside for a follow-up PR.

[jonathan-gramain]

depends on #189 to be merged first, because of the lack of default values for the CI otherwise.