Add regression to the model

[mghosh00]

Allow extra flexibility within the model by allowing continuous labels rather than just categorical ones. This will involve changing/adding the following classes:

components

• [x] Network.__init__
• [x] Network.forward_pass_one_datapoint (else block)
• [x] Network.store_gradient_of_loss (output layer)
• [x] tests

main

• [x] AbstractSimulator.__init__
• [x] AbstractSimulator.forward_pass_one_batch
• [x] AbstractSimulator.abs_generate_scatter (change the method which Plotter calls)
• [x] Trainer.run, Tester.run, Validator.validate (just don't call _update_categorical_dataframe)
• [x] Plotter.comparison_scatter (plot against line y = x)
• [x] tests

functions

• [x] Loss (rename to CrossEntropyLoss)
• [x] MSELoss (remember to just return the squared difference)
• [x] tests