Nested resampling (nested CV) with several metrics using tidymodels

I tried to recreate this example of nested resampling with tidymodels but with two hyperparameters and more than one metric.

Here is the code for two hyperparameters:

# Data
library(mlbench)
sim_data <- function(n) {
  tmp <- mlbench.friedman1(n, sd = 1)
  tmp <- cbind(tmp$x, tmp$y)
  tmp <- as.data.frame(tmp)
  names(tmp)[ncol(tmp)] <- "y"
  tmp
}

set.seed(9815)
train_dat <- sim_data(50)
train_dat$y <- rep(c("yes", "no"))
large_dat$y <- rep(c("yes", "no"))
train_dat$y <- as.factor(train_dat$y)

# Nested CV
library(tidymodels)
results <- nested_cv(train_dat, 
                     outside = vfold_cv(v= 3, repeats = 3), 
                     inside = bootstraps(times = 5))
results

# Apply SVM to nested CV
library(kernlab)

# `object` will be an `rsplit` object from our `results` tibble
# `cost` is the tuning parameter
svm_metrics <- function(object, cost = 1, rbf_sigma = 0.2) {
  y_col <- ncol(object$data)
  mod <- 
    svm_rbf(mode = "classification", cost = cost, rbf_sigma = rbf_sigma) %>% 
    set_engine("kernlab") %>% 
    fit(y ~ ., data = analysis(object))
  
  holdout_pred <- 
    predict(mod, assessment(object)) %>% 
    bind_cols(assessment(object) %>% dplyr::select(y))
  
 sens(holdout_pred, truth = y, estimate = .pred_class)$.estimate
}

# In some case, we want to parameterize the function over the tuning parameter:
svm_metrics_wrapper <- function(cost, rbf_sigma, object) svm_metrics(object, cost, rbf_sigma)


# `object` will be an `rsplit` object for the bootstrap samples
tune_over_svm <- function(object){
  tibble(cost = grid_random(cost(), size = 3),
         rbf_sigma = grid_random(rbf_sigma(), size = 3)) %>% 
    mutate(Sens = map2_dbl(cost, rbf_sigma, svm_metrics_wrapper, object = object))
}

# `object` is an `rsplit` object in `results$inner_resamples` 
summarize_tune_results <- function(object) {
  # Return row-bound tibble that has the 25 bootstrap results
  map_df(object$splits, tune_over_svm) %>%
    # For each value of the tuning parameter, compute the 
    # average sensitivity which is the inner bootstrap estimate. 
    group_by(cost, rbf_sigma) %>%
    summarize(mean_sens = mean(Sens, na.rm = TRUE),
              n = length(Sens),
              .groups = "drop")
}


library(furrr)
plan(multisession)

tuning_results <- future_map(results$inner_resamples, summarize_tune_results) 

However, I am totally at loss how to add other metrics as well, let's say specificity and pr_auc. Any idea how to do this is appreciated. Thanks.