[prometheus-statsd-exporter] fix default liveness/readiness probe path (/health returns 404)

[yuval-s-wiz]

What

The chart's default livenessProbe/readinessProbe use httpGet.path: /health, but statsd_exporter returns HTTP 404 on /health. With the chart's default values the pod fails its probes — readiness never goes Ready, and liveness restarts the container.

Reproduce

$ curl -so /dev/null -w '%{http_code}\n' http://<exporter>:9102/health     # 404
$ curl -so /dev/null -w '%{http_code}\n' http://<exporter>:9102/-/healthy   # 200
$ curl -so /dev/null -w '%{http_code}\n' http://<exporter>:9102/-/ready     # 200

Root cause

/health was introduced in #2305 on the assumption that the server exposes it. It doesn't have a dedicated /health route — statsd_exporter serves /-/healthy and /-/ready (added in prometheus/statsd_exporter#339).

/health returned 200 only incidentally up to statsd_exporter v0.22, whose / handler was a catch-all that returned 200 for any path. In v0.24.0 (2023-06-02) the exporter adopted exporter-toolkit's web.NewLandingPage, which 404s any path != "/" — silently breaking this probe for every statsd_exporter >= v0.24, including this chart's appVersion: v0.28.0 and the upstream prom/statsd-exporter image.

Fix

Point both probes at /-/healthy — the canonical lightweight health endpoint. This also satisfies the original intent of #2305 (avoid probing the heavy /metrics endpoint). Chart version bumped 1.0.0 → 1.0.1.