Inference commands do not allow to specify HyperPod Task Governance configurations

The inference commands to deploy JumpStart model or custom model do not allow to specify task governance configurations.

For example, the commands should support parameters that are similar to what is supported for PyTorch job submission:

```
--scheduler-type TEXT If specified, training job pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
--queue-name TEXT Queue name for job scheduling
--priority TEXT Priority class for job scheduling
--preferred-topology TEXT Preferred topology annotation for scheduling
--required-topology TEXT Required topology annotation for scheduling
```