Methods for keeping object in class

[athowes]

https://github.com/epiforecasts/scoringutils/blob/3d125f12724689cff6de2bf640d4ca992389eff7/R/class-forecast.R#L320

[seabbs]

For others removes the need for asserting the consistency of an object when used which simplifies code quite a bit.

[athowes]

Duplicate of #272.

In #52 I've found that after making some alterations to an object of class latent_individual that I need to recall as_latent_individual in order for epidist::epidist to work:

obs_prep_reduced <- obs_prep |>
  dplyr::group_by(district) |>
  dplyr::mutate(n = dplyr::n()) |>
  dplyr::filter(n > 30) |>
  dplyr::select(-n) |>
  dplyr::ungroup()

obs_prep_reduced <- as_latent_individual(obs_prep_reduced)

This doesn't seem to be ideal. Probably something easy to fix here.

Then @seabbs suggested that it'd been solved in scoringutils (the OP of this post).