• log all the values of the following tables as run metrics: consistency, eval_short, true_score_eval. This allows easy comparison between runs in the runs view

• log confusion matrices as custom charts, this also allows comparing between runs - see in the workspace view

@desilinguist @mulhod The changes make the tutorial run quite slow for me. Please test and let me know what you think.
to test add the following to rsmtool / rsmeval tutorial config (change project name accordingly), and run it in this branch:

    "use_wandb": true,
    "wandb_project": "rsmtool_tutorial",
    "wandb_entity": "etslabs"