Use consistent shapes of the observed proportions

[pawel-czyz]

Currently, we keep the observed proportions either as:

1. An array with shape (n_timepoints, n_variants) or
2. An array with shape (n_variants, n_timepoints).

The problem is that:

1. Different submodules use different convention (e.g., quasimultinomial and plotting submodules).
2. In fact, quasimultinomial submodule uses both conventions simultaneously!

As this is very counterintuitive and can lead to annoying bugs down the line (e.g., when calculating overdispersion), I'm annotating it as bug, rather than refactor.

[dr-david]

Indeed. Should we do n_timepoints t_timepoints ?

[pawel-czyz]

@dr-david I'm working on this (but only on the public API), so that we will have more consistent user experience in our notebooks. (Well, the notebook, actually: we now officially support only frequentist_notebook_jax, but we have some notebooks which have not been officially added yet, but implement very nice prototypes.)

The PR will spread across quite many files (what's not great for reviewing, but I don't see another solution) and I hope it'll be ready on Wednesday.