[FEAT] return losses of TimeGAN training

[isacmoura]

Is your feature request related to a problem? Please describe.
I need to show and use the training losses to do early stopping operations (this could be added to YData library as well) but I saw that during TimeGAN training there is no interface to retrieve this information.

Describe the solution you'd like
During the training step, the losses of the Generator and Discriminator could be returned for each training step. I saw at the code that these losses are captured, but they aren't returned.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.