Currently it exists as Dev-Preview we need to bring it to GA

from https://github.com/stolostron/stolostron/tree/main/dev-preview#observability-instance-sizes

This document summarizes a new feature called Observability Instance Sizes for controlling the scale of your ACM Observability instance. 

Key Feature Summary

The new Instance Sizes feature simplifies the process of scaling MultiClusterObservability (MCO) resources by allowing users to select a predefined size instead of manually configuring resource requests for individual components (like Thanos, AlertManager, etc.) using AdvancedConfig.

• Simplification: Users can now set the scale using a single field, InstanceSize, in the MultiClusterObservability (MCO) Custom Resource (CR).

• Proportional Scaling: Each instance size corresponds to a set of proportionally sized resource requests across all Observability components.

• Supported Sizes: A range of sizes are supported, representing a linear scale of usage: minimal, default, small, medium, large, xLarge, 2xLarge, and 4xLarge.

• Default Setting: The default size is default.

• Opt-in and Overrides: This feature is entirely opt-in. The existing AdvancedConfig settings will always override any InstanceSize configuration.

Instance Size Estimates

The instance sizes proportionally scale the Observability components based on an estimated practical scale of usage, primarily correlating with the number of active timeseries supported.