Transposing labels on a continuous scale [waterfall graph, R]

I am customizing a waterfall graph and am almost at the end result, except for the labels on what is now the x-axis (after coord_flip()). My now x-axis simply sums to 100% and, while using scale_y_reverse() after coord_flip() has helped me keep the bar hierarchy intended, it set the scale at limits = c(100, 0) instead of what I wanted to still keep limits = c(0, 100). This is the correct way to handle it, except that for my use it does not matter as I am only trying to visualize increments in percentages.

Here is a reprex:

data_fig <- data.frame(class = rev(c("Class A", "Class B", "Class C", "Other Classes")),
                       values = rev(c(15, 25, 50, 10)))

waterfalls::waterfall(
    .data = data_fig,
    rect_width = 0.7, 
    fill_by_sign = FALSE,
    put_rect_text_outside_when_value_below = FALSE,
    rect_border = NA,
    total_rect_border_color = NA,
    draw_axis.x = "none") +
    coord_flip() +
    scale_y_reverse()

What I have tried:

• setting limits = (0, 100) in the scale_y_reverse(); rows are removed.
• set axis.text.x = element_blank() and tried overwriting manually, but error related to continuous scale.
• playing with the order of data, coord_flip() and scale_xy_reverse() in different ways.
• tried building a waterfall plot with geom_rect()/geom_segment() but somehow can't get around this specific issue (also as I understood waterfalls lib uses these functions).

I appreciate any ideas of how to keep limits = (0, 100) and the remaining of the graph elements as is. Thank you very much in advance!

Specifying breaks and labels in reverse order to each other seems to generate the result you are after:

waterfalls::waterfall(
  .data = data_fig,
  rect_width = 0.7, 
  fill_by_sign = FALSE,
  put_rect_text_outside_when_value_below = FALSE,
  rect_border = NA,
  total_rect_border_color = NA,
  draw_axis.x = "none") +
  coord_flip() +
  scale_y_reverse(breaks=c(100, 75, 50, 25, 0), labels=c(0, 25, 50, 75, 100))