Fix backtransformation of intercept in stan_surv when prior_PD = TRUE

[sambrilleman]

Intended to fix #479 for stan_surv models at least.

[jgabry]

Looks good to me!

[sambrilleman]

I'll hold off on merging this until we finish the discussion in the open issue thread (#479 and/or #480, which are each slight variations on a similar problem). I would want to keep stan_surv consistent with whatever happens in the other rstanarm modelling functions I guess.

[sambrilleman]

I'm going to close this unmerged, taking @bgoodri's comments in https://github.com/stan-dev/rstanarm/issues/479 as the reason to not merge:

The recentering always happens because the draw from the prior distribution of the coefficients is never exactly zero, although it is never far from zero under the default prior. I would have preferred no conditioning on the outcome variable at all, but Andrew forced us to for the book. I also agree that stan_surv is weirder because the prior predictive distribution it draws from is even less prior than for the other stan_ functions, but I think the way it deals with the coefficients is fine.

But if anyone feels strongly about it going the other way, we could reopen and merge.