Fix helm metrics prefix issue

[andy-g]

When deploying the Prometheus NATS exporter via the standard Helm charts, the metrics prefix is overridden from gnatsd & jetstream to nats.

This has caused issues where using the standard Jetstream dashboard doesn't work on a standard helm deployment. Please see the related conversation here: #155.

On a partially related note, I was also wondering if it's be worth adding a simple docker-compose.yaml file in the walkthrough folder, so anyone interested could easily get a NATS + Prometheus + Prometheus NATS Exporter + Grafana environment up and running. The only concern I have is that we've configured the dashboards as templates to be manually imported (which includes the __inputs & __requires sections), as opposed to exported dashboards to automatically import (which would include the actually values for the prometheus datasource). We could either leave it to the user to still import the dashboard, or include a subfolder with the required configurations for the docker-compose example. Let me now if there's any interest in this and I can spin up a separate PR.

[variadico]

I was also wondering if it's be worth adding a simple docker-compose.yaml file in the walkthrough

Yeah, that sounds like a good idea.