Include latent variables

[ScottClaessens]

Currently, the package does not allow for the inclusion of latent variables. I was thinking that a good way to implement this within our current architecture would be with nested lists. So instead of the case of separate variables

list(
  x = "bernoulli_logit",
  y = "ordered_logistic
)

we would have a nested list indicating that x is a latent variable made up of three manifest variables

list(
  x = list(
    x1 = "bernoulli_logit",
    x2 = "normal",
    x3 = "student_t"
  ),
  y = "ordered_logistic"
)

This isn't a priority, but probably something to think about for future iterations of the package.