Needs validation as a user story. Models have input token limits, so having limits in 2 places is questionable. Also, input tokens are included in token usage in response already for TokenRateLimitPolicy as of v1alpha1

As a platform engineer, I want to enforce input token rate limits at the Gateway and HTTPRoute level so that I can prevent excessive usage of expensive LLM APIs before requests reach the model server.