Log artifact in folders

[icedoom888]

Is your feature request related to a problem? Please describe.

Currently when logging using MLflow, all artifacts are saved at the same level.
This results in an extremely dense and hard to navigate list of saved plots.

Describe the solution you'd like

Use MLFlow artifact folders for each Callback:

import mlflow

with mlflow.start_run():
    # This logs 'my_plot.png' into a folder named 'visuals'
    mlflow.log_artifact("my_plot.png", artifact_path="visuals")
    
    # This logs 'config.yaml' into a nested folder 'config/hyperparams'
    mlflow.log_artifact("config.yaml", artifact_path="config/hyperparams")

This will make looking at MLFlow artifacts easier and more effective.

Describe alternatives you've considered

No response

Additional context

No response

Organisation

Meteoswiss