gamma_dist_def: scale is inverse scale?

[seabbs]

It looks like in gamma_dist_def that scale is actually its inverse? This matches with the stan parameterization (which was passed in some models as alpha beta) but not with the R version.

https://github.com/epiforecasts/EpiNow2/blame/961c8c91063c6da381633ada387e2fba0337c101/R/dist.R#L340

THe renaming slipped through in the recent work on dist_spec but I am not clear if there was a deeper functionality issue and if that remains.

[sbfnk]

True but then it's passed (positionally) as inverse scale in

EpiNow2/R/dist.R

Line 102 in af8c1f3

rgamma(n, params$shape, params$scale)

Both should probably be fixed as part of #382