removed all partial imports and all import * with the aim of avoiding potential circular imports issues and increase readability. Fixed a couple bugs along the way

Manually bringing all the mkdocs changes to the self supervised branched as the conflicts were somehow too much to make it happen via sublime merge or vscode...I am at utils. After this we need to test callbacks for self supervised. Then the usual overall checks and install checks and make the massive release

Added Self Supervised Pretraining for all tabular models. Need to unit test and surely there will be some code readjustments

Self Supervised runs for 4 attention-based models. The Perceiver will not be supported and not need to test the MLP-ATtention models

Self Supervised seems to be running in a small number of cases. Now I need to test a few more cases. Adjust unit test and re-organise code

Self Supervised Runs for all attention based models. Whether it learns something or not, is a question we still need to answer

Adjusted output and encoder_output dims so that all models can be used with self supervised training

Added base trainers for the Trainer and the BayesianTrainer classes. The base trainers will change as I add self supervised options

adjusting the types so that the library is compatible with the newer version of torch 1.11 and torchvision 0.12

* reducelronplateau implementation fixed + added test
* added test for _extract_kwargs() in trainer
* added additional coverage for fds_layer
* added lds weight test coverage for losses