Fixes for Laplace, Logistic and Weibull

[MichalLauer]

Proposed changes

This PR fixes several bugs within distributions.

Types of changes

Put an x in the boxes that apply or remove the lines that don't.

• Bugfix (non-breaking change which fixes an issue)

Checklist

Put an x in the boxes that apply or remove the lines that don't. This is a reminder of what we are going to look for before merging your code.

• I have added unit tests and all tests pass locally with my changes
• I have added tests that prove my fix is effective or that my feature works
• I have added documentation where necessary
• Any dependent changes have been merged and published in downstream modules

Further comments

Fixes the following errors:

# Laplace distribution
Laplace$new(var = 4)
# Logistic distribution
Logistic$new(sd = 1)$getParameterValue('mean')
Logistic$new(sd = 1)$getParameterValue('scale')
Logistic$new(sd = 1)$getParameterValue('sd')
# Weibull distribution
Weibull$new(altscale = 3)$getParameterValue("shape")
Weibull$new(altscale = 3)$getParameterValue("scale")
Weibull$new(altscale = 3)$getParameterValue("altscale")

[RaphaelS1]

Thanks @MichalLauer. The first commit (359a137) is unreviewable, I can't tell what's been changed there at all. In the next two I can see the variable orders are reversed. Please let me know what you changed in the first commit

[MichalLauer]

Hey @RaphaelS1, apologies, that's strange. I will look into it and try to fix the commit so it's clear what I've changed. For clarity, the change is in the computation of scales.

getParameterSet.Laplace <- function(object, ...) {
  pset(
    prm("mean", "reals", 0, tags = "required"),
    prm("scale", "posreals", 1, tags = c("linked", "required")),
    prm("var", "posreals", tags = c("linked", "required")),
    trafo = function(x, self) {
        scales <- list_element(x, "scale")
        vars <- list_element(x, "var")
        if (length(scales) && length(vars)) {
          stop("Can't update 'scale' and 'var' parameters simultaneously")
        }
        if (length(scales)) {
          vars <- setNames(2 * unlist(scales)^2,
                           gsub("scale", "var", names(scales)))
        } else {
         # !!! Here added unlist so it does not throw an error
          scales <- setNames(sqrt(unlist(vars) / 2),
                             gsub("var", "scale", names(vars)))
        }

        unique_nlist(c(vars, scales, x))
    }
  )
}