Grafana k8s Monitoring (Helm Chart)

What's changed

• Fix the deployment notes printing an empty "It will:" section when no local features are enabled (e.g. when only using remote configuration or only deploying telemetry services) (#2711) (@petewall)
• Fix the OTLP destination timeout setting rendering timeout at the top level of the otelcol.exporter.otlp/otelcol.exporter.otlphttp component instead of inside the client block. Alloy 1.16.3+ rejects the misplaced attribute (unrecognized attribute name "timeout"), causing the collector to fail to start. (#2710) (@petewall)
• Update Alloy Operator to 0.5.10 and kube-state-metrics to 7.5.1 (@petewall)
• Expand the platform detection template to recognize Azure AKS, Google GKE, and Amazon EKS (in addition to OpenShift) by inspecting node labels and cloud provider IDs, and expose the detection result through a reusable validations.platform.detect template. (@petewall)
• Report the detected platform in the platform label of the grafana_kubernetes_monitoring_build_info self-reporting metric when global.platform is not set explicitly. (@petewall)* Set the provider attribute reported by the Remote Configuration (remotecfg) feature to the configured or detected platform (openshift, aks, gke, eks) when it can be determined. (@petewall)

Full changelog: CHANGELOG.md