priors module added

[u-dupletsa]

priors module (tuturial coming soon)

[jacopok]

Hi, some comments:

• the code to apply these priors to an event should be part of the module, I think, not only as a tutorial - the user should ideally be able to apply priors with a simple function call
• the paper by Botev mentioned here should be cited (I cannot find it in the overleaf, maybe I missed it)
• the test code in minimax_tilting_sampler.py should be made into an actual test, I think, also
• it does not seem to give sensible results: this is what I get (adding a np.random.seed(10) for reproducibility on line 432, and nothing else:

• also, the code there is not super convincing to me: you're plotting on two separate axes obtained with twinx, so normalization is lost - the distributions should be correctly normalized, either multiplying the PDF by the number of samples, or by using the density=True kwarg for the histogram