[Feature Request] Return depth from RoboHiveEnv

[sriramsk1999]

Motivation

Simulated environments are capable of rendering depth images which can be very useful in downstream training as they enable 3D understanding of the task being performed. RoboHive environments, simulated with mujoco, can render and return depth images along with RGB. However, this functionality is currently not exposed in the torchrl wrapper RoboHiveEnv.

Solution

A flag enabling depth images to be returned from RoboHiveEnv. The solution has been implemented here

Checklist

• I have checked that there is no similar issue in the repo (required)