## Check for existing issues

- [x] I searched the existing issues and checked that this is not a duplicate

## What happened?

When Azure OpenAI image generation or image editing is configured with `api_version: v1` or `preview`, LiteLLM Proxy builds a deployment-scoped URL instead of the integrated Azure v1 image endpoint

The request is sent to `/openai/deployments/{model}/images/generations` or `/openai/deployments/{model}/images/edits` instead of `/openai/v1/images/generations` or `/openai/v1/images/edits`. This can result in provider errors such as `404 Resource not found`

This affects the Azure OpenAI provider through the LiteLLM Proxy

## Steps to Reproduce

1. Configure an Azure OpenAI image deployment in LiteLLM Proxy with `api_version: v1` or `preview` 2. Call `POST /v1/images/generations` or `POST /v1/images/edits` 3. Inspect the upstream request URL 4. Observe that LiteLLM uses the deployment-scoped route instead of the integrated v1 route

## Relevant log output

```text Expected: /openai/v1/images/generations or /openai/v1/images/edits Actual: /openai/deployments/{model}/images/generations or /openai/deployments/{model}/images/edits ```

## What part of LiteLLM is…