✨ stats: multinomial shape-typing

[jorenham]

towards #406

ref: #1549 (review)

[jorenham]

cc @Aniketsy; this turned out to be way more complex that I had hoped... And since I don't expect it to be any simpler for the other multivariate distributions, it's absolutely fine (and maybe even wise) if you don't feel like working on these multivariate distributions.

[github-actions]

mypy_primer results

✅ No ecosystem changes detected

[Aniketsy]

cc @Aniketsy; this turned out to be way more complex that I had hoped... And since I don't expect it to be any simpler for the other multivariate distributions, it's absolutely fine (and maybe even wise) if you don't feel like working on these multivariate distributions.

thanks, yes i can see quite large changes, so i'll not touch these as diving into these will create extra work for the review, but i'll try to understand how it is done to have some knowledge, and maybe in near future i can touch these ...
for now i'll try to pick simple and moderate level issues 😁(is this good way to move towards perfectionist ? If not please correct me) , also please feel free to ping me if you feel any issue fits for me.