feat(chart): add configurable ServiceMonitor support

[somaz94]

Adds opt-in Prometheus Operator ServiceMonitor support to the supabase chart, addressing #137.

A top-level serviceMonitor block gates creation behind a master switch plus a per-component toggle. A ServiceMonitor is created only for a component that is both deployed (deployment.<component>.enabled) and enabled under serviceMonitor.components.<component>. Global defaults (interval, scrapeTimeout, port, path, scheme, tlsConfig, relabelings, metricRelabelings) can be overridden per component. Everything defaults to off, so the rendered output is unchanged unless a user opts in, and no Prometheus Operator CRD is required for the default install.

To let each ServiceMonitor select a single component's Service, the recommended app.kubernetes.io/component label is added to every component Service. This is additive and does not change any existing selector.

Chart version bumped 0.6.0 -> 0.7.0 (minor, new feature).

Validation (local):

• helm lint and ct lint pass
• helm template with default values renders 0 ServiceMonitor
• helm template with serviceMonitor.enabled plus per-component toggles renders one ServiceMonitor per enabled and deployed component, with per-component overrides applied
• a component whose deployment is disabled renders no ServiceMonitor
• ct install was not run locally (it needs a kind cluster). Because the feature defaults to off, the installed resources are unchanged apart from the additive Service label.

related: #137

[somaz94]

Hi @luizfelmach — could you approve the workflow runs for this PR?

No checks have executed yet; they are queued awaiting first-time-contributor approval, so I have no way to show the change is green.

The PR adds configurable ServiceMonitor support to the chart (opt-in, disabled by default) and bumps the chart version. Thanks!

[luizfelmach]

Hi @somaz94. Thanks for that. I need some time to test this.