Option to track CPU/GPU metrics

MLFlow can track the compute metrics similar to WandB. It would be good to give the user the possibility to activate it.

See https://mlflow.org/docs/latest/system-metrics/index.html

The metrics can be logged with:
```
logger = MLFlowLogger(
experiment_name=LOG_EXPERIMENT_NAME,
run_name=run_name,
tracking_uri=f"file:{LOG_DIR}{LOG_PROJECT}",
)
logger.log_hyperparams(config)

# Start logging system metrics
mlflow.start_run(
run_id=logger.run_id,
experiment_id=logger.experiment_id,
log_system_metrics=True)
```