Is your enhancement related to a problem? Please describe

As a user, I should be able to configure the inferencing runtime when starting a playground service. It should be one more option, using a select box as we do for the model.

Also, the playground environment is using the existing model service, so we'll need to be careful about that. One way to handle it, would be to show the existing model services that are already running. I believe a "Model Service" section would be needed in the form. That section would be displaying the combination between the model and the inferencing runtime - this way, the user could select an already running one, or alternatively create a new one.

Details:

• As a user, I should be able to identify the default runtime
• As a user, I should be able to start multiple model services with the same configuration (model + inferencing runtime) using different ports
• As a user, I should be able to know if the inferencing runtime image is already pulled locally
• As a user, I should not be proposed the inferencing runtime if it is not compatible with my environment

Describe the solution you'd like

Similar pattern than in the UX ticket: #2502

Describe alternatives you've considered

No response

Additional context

No response