Is there an existing issue for this?

•  I have searched the existing issues

Deploy type

OpenDataHub core version (eg. v1.6.0)

Version

2.4

Current Behavior

Due to how we fetch the Servingruntimes and Inferenceservices to avoid issues with permissionless users, clusters with a lot of projects will impact the performance of the request.

Expected Behavior

Users listing the resources with less performance impact.

Steps To Reproduce

1. Create several projects
2. Deploy models in those projects
3. Maybe creat throttle in the network inspector (to highlithg the issue)
4. Go to model serving page