[bug] cell type proportions approach doesn't correct for multiple testing

[Al-Murphy]

[Al-Murphy]

Dirichlet approach here doesn't appear to be corrected for multiple testing.

I think the function .process_dirichlet_fit() should be adjusted for multiple testing before adding the significance markers.

Maybe add a hyperparameter for the user to decide which cells they want to test for different cell type proportions (e.g. if I'm only interested in microglia I don't want to decrease power by checking proportions in other cell types). By default though this should be all cell types.

Do let me know if I'm missing where this is being done although I don't see it