Apply a function across all combinations of variables and store the plot in a list where we can plot all (or part) of the ggplot lists

I have some data (iris dataset) and I also have a plotting function called decisionplot. My question is how can I apply the decisionplot function (which outputs a ggplot plot) to all combinations of variables in my data. That is, in the iris data set we have;

"Sepal.Length""Sepal.Width""Petal.Length""Petal.Width""Species"

I want to apply the function decisionplot to the following combinations:

"Sepal.Length""Sepal.Width""Sepal.Length""Petal.Length""Sepal.Length""Petal.Width""Sepal.Width""Petal.Length""Sepal.Width""Petal.Width""Petal.Length""Petal.Width"

Store these plot combinations in a list where I can plot on a grid the different outputs.

The data and the decisionplot function is the following:

data(iris)
df <- iris %>% 
  filter(Species != "setosa") %>% 
  mutate(Species = +(Species == "virginica"))


decisionplot <- function(model, data, class = NULL, model_type = "NA", predict_type = "class",
                         resolution = 300, main, showgrid = TRUE, ...) {

  if(!is.null(class)){
    cl <- data[,class]
  }
  else{
    cl <- 1
  }

  k <- length(unique(cl))



  # make grid
  r <- sapply(data[,1:2], range, na.rm = TRUE)
  xs <- seq(r[1,1], r[2,1], length.out = resolution)
  ys <- seq(r[1,2], r[2,2], length.out = resolution)
  g <- cbind(rep(xs, each=resolution), rep(ys, time = resolution))
  colnames(g) <- colnames(r)
  g <- as.data.frame(g)

  ### guess how to get class labels from predict
  ### (unfortunately not very consistent between models)
  if(model_type == "xgboost"){
    message(paste0("Running Model ", model_type))
    p <- predict(object = model, newdata = xgboost::xgb.DMatrix(data = as.matrix(g)), type = predict_type)
    #p <- ifelse(p > tail(sort(p),1000)[1000], "Bankrupt", "Non-Bankrupt")
    p <- ifelse(p > 0.50, "Bankrupt", "Non-Bankrupt")
  }
  else {
    message(paste0("Running Model ", model_type))
    p <- predict(model, g, type = predict_type)
  }
  if(is.list(p)) p <- p$class
  if(is.logical(p)) p <- ifelse(p == TRUE, "Non-Bankrupt", "Bankrupt")
  p <- as.factor(p)

  plot_data <- cbind(g, p) %>% 
    setNames(c("X1", "X2", "Y")) %>% 
    mutate(class_num = case_when(
      Y == "Non-Bankrupt" ~ 1,
      Y == "Bankrupt" ~ 2
    ))
  ggplot() +
    geom_point(aes(x = X1, y = X2, colour = Y),
               data = plot_data) +
    geom_contour(aes(x = X1, y = X2, z = class_num),
                 bins = 2,
                 data = plot_data, size = 0.001) +
    geom_point(aes(x = Xvar1, y = Xvar2, color = factor(Yvar)),
               size = 3,
               data = data) +
    geom_point(aes(x = Xvar1, y = Xvar2),
               size = 3,
               shape = 1,
               #alpha = 0.2,
               data = data) +
    labs(title = paste(main),
         x = var1_to_plot,
         y = var2_to_plot)

}

I then set my variable I want to put through the function and create my boundary_data:

var1_to_plot = "Sepal.Length"
var2_to_plot = "Sepal.Width"

boundary_data <- df %>% 
  dplyr::select(rlang::eval_tidy(var1_to_plot), rlang::eval_tidy(var2_to_plot), Species) %>% 
  mutate(Species = factor(Species)) %>% 
  setNames(c("Xvar1", "Xvar2", "Yvar")) %>% 
  data.frame()

I finally create my logistic model and run the decisionplot function.

model <- glm(Yvar ~., data = boundary_data, family = binomial(link='logit'))
class(model) <- c("lr", class(model))
predict.lr <- function(object, newdata, ...)
  predict.glm(object, newdata, type = "response") > .5
decisionplot(model, boundary_data, predict_type = "response", class = "Yvar", model_type = "Logistic", main = "Logistic Regression")

This gives me:

However I want to make it such that this will be one of all the combinations of the variables (since here it just consideres Sepal.Width and Sepal.Length).

How can I plot on a grid the different ggplot variable combinations? Or store the plots as a list?

EDIT

What I currently have is the following - Which gives me 12 lists of different combinations. Each list has 2 further lists a data set and a logistic model.

var_combos <- expand.grid(colnames(df[,1:4]), colnames(df[,1:4])) %>% 
  filter(!Var1 == Var2)


iter_function = function(dat, V1, V2, Y){
  data = dat %>% 
    select(rlang::eval_tidy(V1), rlang::eval_tidy(V2), Y) %>% 
    mutate(Y = factor(rlang::eval_tidy(Y)))

  # Logistic Model
  model = glm(rlang::eval_tidy(Y) ~ ., data = data, family = binomial(link = 'logit'))
  class(model) <- c("lr", class(model))
  predict.lr <- function(object, newdata, ...){
    predict.glm(object, newdata, type = "response") > .5
  }
  return(list(data, model))
}

boundary_lists <- map2(
  .x = var_combos$Var1,
  .y = var_combos$Var2,
  ~iter_function(dat = df, V1 = .x, V2 = .y, Y = df$Species)
)

Now I want to apply each of these lists to the decisionplot function. I keep getting . not found.

map2(
  .x = .[[1]],
  .y = .[[2]],
  ~decisionplot(model = .x, data = .y, predict_type = "response", class = "Species", model_type = "Logistic", main = "Logistic Regression")
)

How can I map the funtion of list of lists? (I realise I am not giving it the boundary_lists list created from the iter_function())