multi-predict for some prediction types

[topepo]

With glmnet (and maybe mboost) we can get predictions one many values of some tuning parameters without multiple model fits. parsnip saves those as nested tibbles per row of new_data.

This is also how we store the predictions of hazard and survival probabilities.

We might be able to accommodate both by having a single nested for both. For example, for glmnet, we would have the full grid of penalty and .time values with an additional column for the prediction.

[hfrick]

#70 covers multi_predict() for glmnet's Cox models for the prediction types "survival" and "linear_pred".

[hfrick]

todo: type = "time" for glmnet

[hfrick]

possible todo: mboost

[hfrick]

Added the remaining prediction types of glmnet models in #282 , and am going to close this issue in favor of a new one specifically for mboost: #290

[github-actions]

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.