Graduate InferenceModelRewrite to v1 (GA)

[zetxqx]

This is a tracking issue for graduating the inferenceModelRewrite CRD to v1 (General Availability).

The following list of tasks tracks the requirements for graduation. This list is subject to updates as the feature evolves.

API & Stability

• API Review: Complete formal API review for the InferenceModelRewrite CRD to ensure schema stability and adherence to API conventions.

Documentation

• User Guide: Create a dedicated guide explaining how to configure and use Model Rewrite and Traffic Splitting.

Observability & Operations

• Metrics: Implement Prometheus metrics to track traffic splitting and model rewrites
• Logging: Ensure clear debug/info logs are emitted when a request is rewritten, including the original and target model names.