Fix error message for multi_predict()

[juliasilge]

Closes #482

This PR fixes the error message for multi_predict() and adds a few small tests for the error message:

library(tidymodels)
#> Registered S3 method overwritten by 'tune':
#>   method                   from   
#>   required_pkgs.model_spec parsnip
data(attrition)

data_split <- initial_split(attrition, strata = "Attrition")
attrition_train <- training(data_split)
attrition_test  <- testing(data_split)

tree_mod <-
  boost_tree(mode = "classification", trees = 100) %>% 
  set_engine("C5.0") %>% 
  fit_xy(x = attrition_train %>% dplyr::select(-Attrition), y = attrition_train$Attrition)

multi_predict(
  tree_mod, 
  new_data = attrition_test %>% dplyr::select(-Attrition),
  trees = 80:100,
  type = "prob"
)
#> # A tibble: 367 x 1
#>    .pred            
#>    <list>           
#>  1 <tibble [21 × 3]>
#>  2 <tibble [21 × 3]>
#>  3 <tibble [21 × 3]>
#>  4 <tibble [21 × 3]>
#>  5 <tibble [21 × 3]>
#>  6 <tibble [21 × 3]>
#>  7 <tibble [21 × 3]>
#>  8 <tibble [21 × 3]>
#>  9 <tibble [21 × 3]>
#> 10 <tibble [21 × 3]>
#> # … with 357 more rows

glm_mod <-
  logistic_reg(mode = "classification") %>% 
  set_engine("glm") %>% 
  fit(Attrition ~ ., data = attrition)

multi_predict(
  glm_mod, 
  new_data = attrition_test %>% dplyr::select(-Attrition),
  type = "prob"
)
#> Error: No `multi_predict` method exists for objects with classes '_glm', 'model_fit'

^{Created on 2021-05-06 by the reprex package (v2.0.0)}

[topepo]

💯

[github-actions]

This pull request 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.