added generalized gamma distribution

[kylejcaron]

What is this PR about? This PR implements the generalized gamma distribution, using the parameterization GG(alpha, p, lambda).

This corresponds with the following PR in aesara.

It also was originally this PR in pymc, but @ricardoV94 recommended to start off with pymc-experimental which makes sense to me especially since I'm having trouble using it with the pm.Censored API

[ricardoV94]

@OriolAbril Can you give some guidance on how to include the distribution in the doc pages?

[OriolAbril]

It needs to be added to https://github.com/pymc-devs/pymc-experimental/blob/main/docs/api_reference.rst. Which also needs some extra changes (along with maybe some changes to the imports and module hierarchy) I think. For this PR only:

:mod:`pymc_experimental.distributions`
=============================

.. automodule:: pymc_experimental.distributions
   :members: histogram_approximation, GeneralizedGamma

In general, is the plan for users to use pymc_experimental.utils.prior.prior_from_idata or to use pymc_experimental.utils.prior_from_idata? Depending on what is the plan, the imports and structure when generating the API docs should be updated. The docs currently tell users to import prior_from_idata from pmx.utils.prior