Transform long format dataframe with multiples sets of columns into wide format

Question

How can you transform a long format dataframe with multiple sets of columns into a wide format?

df <- data.frame(
  id = c(1, 2),
  uid = c("m1", "m2"),
  germ = c(23, 24),
  category = c("x1", "x2"),
  mineral = c(78, 10))

Names come from uid and category while Values come from germ and mineral. I have tried working on this with the code below but the names are not what I want

out_df <- df %>% 
  tidyr::pivot_wider(names_from = c(uid, category), values_from = c(germ, mineral))

Here is what I would like my output to look like

out_df <- data.frame(id = c(1, 2),
                     m1 = c(23, NA),
                     m2 = c(NA, 24),
                     x1 = c(78, NA),
                     x2 = c(NA, 10))