Add vignette on computation with general likelihoods

[pcarbo]

ashgen.Rmd.gz

Thanks to Mengyin.

[stephens999]

@mengyin I think it would be good to simplify the use of the general likelihoods by setting betahat=0 and betahatsd=1 for the user when using likelihoods that require that.

In fact, maybe that should be the default: if a likelihood is specified and not a t or normal likelihood, and betahat and bhatsd are missing, set betahat=0 and the sd=1.

So for example ash(rep(0, length(y)), 1, lik = lik_pois(y)) would become just ash(lik = lik_pois(y))

What do you think?

[mengyin]

I agree! Having betahat/sebetahat for the general likelihoods indeed looks confusing. It will be much more clear if we can just avoid them.