[RLlib] - `MultiAgentEnvRunner` misses multi-agent metrics

[simonsays1980]

What happened + What you expected to happen

What happened

I ran the tuned example [multi_agent_pendulum_ppo_envrunner.py](https://github.com/ray-project/ray/blob/master/rllib/tuned_examples/ppo/multi_agent_pendulum_ppo_envrunner.py) and did not see single agent metrics in Tensorboard. Debugging showed, they are not collected.

What you expected to happen

That the new EnvRunner API collects single agent metrics in a multi-agent setting like the new Learner API does.

Versions / Dependencies

Ray nightly (Feb. 17th) Linux Fedora 37 Python 3.11.0

Reproduction script

multi_agent_pendulum_ppo_envrunner.py

Issue Severity

Medium: It is a significant difficulty but I can work around it.