Making use of LightningDataModule and simplification of data loading

Closes #10, #8, #15, #16, #27, #32, #28, #44, #29, #21, #19, #18, and #14

See merge request biosignal/software/ptbench!6

Since lightning 2.1.0, the error "The classmethod `Pasa.load_from_checkpoint`
cannot be called on an instance." was returned when this method was called,
making the tests fail.

This reverts commit 20dfcdb4.

[test/test_cli] updated unit tests for training, prediction and evaluation cli scripts. Removed radiological signs (rs) related tests

[scripts.experiment] Reflect changes from evaluation; closes #44 after noticing train-analysis is performed

[engine.predictor] Allow prediction to work when batch-size!=0; Discard CSVs and use JSON for output; Implement lightning logging like in training; Separates batched (collated) data at prediction; Adapts all models to new paradigm; Remove CSVPredictionsWriter; Adapt predict script to all changes