Fixes/inference tissue subtypes

[AjeyPaiK]

This PR:

1. Fixes{#1, #2}
2. Adds default data description configs for tissue subtype segmentation.
3. Adds a flag that allows logging epoch level loss while training.
4. Adds default configs for inference trackers.

[jonasteuwen]

Early comments:

• Please remove all data_descriptions that are specific for our use-cases
• Why are there a bunch of files that seem edited, to me it appears as if nothing happened there?
• Logging the loss per epoch, does this also log when we evaluate more often than once per epoch? Perhaps we can have both (epoch loss, and batch loss?)
• Your edit to disable the mask, how will that work when we have a tissue-bg mask?

[AjeyPaiK]

1. I have removed the configs we use for our internal experiments.
2. Some files seem edited even though nothing happened due to Black. Now, this repo passes black.
3. We have both epoch level loss and also batch level loss being logged. Although, the batch level loss will be logged once every 50 steps (which is the pytorch lightning default).
4. Yes, better solutions are possible rather than setting mask to None during inference. Perhaps a separate manifest for inference? Added it as a fixme.

[jonasteuwen]

I agree with the changes for now