Fixes distributed reporting for RSL RL benchmark (#498)

# Description

RSL RL reports data for a single GPU. For multi-GPU benchmarking, we
multiple the data by the number of GPUs to capture the total FPS.


## Type of change

<!-- As you go through the list, delete the ones that are not
applicable. -->

- Bug fix (non-breaking change which fixes an issue)


## Checklist

- [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with
`./isaaclab.sh --format`
- [x] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] I have added tests that prove my fix is effective or that my
feature works
- [ ] I have updated the changelog and the corresponding version in the
extension's `config/extension.toml` file
- [ ] I have added my name to the `CONTRIBUTORS.md` or my name already
exists there

<!--
As you go through the checklist above, you can mark something as done by
putting an x character in it

For example,
- [x] I have done this task
- [ ] I have not done this task
-->

isaaclab.sh

@@ -390,6 +390,7 @@ while [[ $# -gt 0 ]]; do
             if ! command -v pre-commit &>/dev/null; then
                 echo "[INFO] Installing pre-commit..."
                 pip install pre-commit
+                sudo apt-get install -y pre-commit
             fi
             # always execute inside the Isaac Lab directory
             echo "[INFO] Formatting the repository..."

scripts/benchmarks/benchmark_rsl_rl.py

@@ -143,6 +143,7 @@ def main(env_cfg: ManagerBasedRLEnvCfg | DirectRLEnvCfg | DirectMARLEnvCfg, agen
 
     # multi-gpu training configuration
     world_rank = 0
+    world_size = 1
     if args_cli.distributed:
         env_cfg.sim.device = f"cuda:{app_launcher.local_rank}"
         agent_cfg.device = f"cuda:{app_launcher.local_rank}"
@@ -152,6 +153,7 @@ def main(env_cfg: ManagerBasedRLEnvCfg | DirectRLEnvCfg | DirectMARLEnvCfg, agen
         env_cfg.seed = seed
         agent_cfg.seed = seed
         world_rank = app_launcher.global_rank
+        world_size = int(os.getenv("WORLD_SIZE", 1))
 
     # specify directory for logging experiments
     log_root_path = os.path.join("logs", "rsl_rl", agent_cfg.experiment_name)
@@ -221,13 +223,17 @@ def main(env_cfg: ManagerBasedRLEnvCfg | DirectRLEnvCfg | DirectMARLEnvCfg, agen
 
         # prepare RL timing dict
         collection_fps = (
-            1 / (np.array(log_data["Perf/collection time"])) * env.unwrapped.num_envs * agent_cfg.num_steps_per_env
+            1
+            / (np.array(log_data["Perf/collection time"]))
+            * env.unwrapped.num_envs
+            * agent_cfg.num_steps_per_env
+            * world_size
         )
         rl_training_times = {
             "Collection Time": (np.array(log_data["Perf/collection time"]) / 1000).tolist(),
             "Learning Time": (np.array(log_data["Perf/learning_time"]) / 1000).tolist(),
             "Collection FPS": collection_fps.tolist(),
-            "Total FPS": log_data["Perf/total_fps"],
+            "Total FPS": log_data["Perf/total_fps"] * world_size,
         }
 
         # log additional metrics to benchmark services