Horizontally match levels of Likert chart based on another bar plot in R

i have a data frame in r called df similar to these question found here:

set.seed(123)
likert_levels <- c(
  "1" = "Very  Dissatisfied",
  "2" = "Dissatisfied",
  "3" = "Neutral",
  "4" = "Satisfied",
  "5" = "Very  Satisfied")
df = data.frame(
  year = sample(c(2023, 2022, 2020, 2018), 50, replace = TRUE),
  A = sample(likert_levels, 50, replace = TRUE),
  B = sample(likert_levels, 50, replace = TRUE),
  C = sample(likert_levels, 50, replace = TRUE))%>%
  mutate(across(everything(),as.factor))%>%
  as_tibble()%>%
  mutate(across(-year, ~ factor(.x, levels = likert_levels)))
df

the likert chart i can do it like this :

mutate(df, id=row_number()) |>
  pivot_longer(-c(id, year), names_to="group") |>
  pivot_wider(names_from=year) |>
  gglikert(c(`2023`, `2022`, `2020`, `2018`), 
         facet_rows=vars(group))

but in this setup i want to sort the likert plot based on the descending sorting of the bar plot . I.e i want to sort first the bar plot in a descending order and then based on this sorting to be sorted accordingly the likert plot . How can i do it in R ?

Here is one potential approach:

library(tidyverse)
library(ggstats)
library(patchwork)

set.seed(123)
likert_levels <- c(
  "1" = "Very  Dissatisfied",
  "2" = "Dissatisfied",
  "3" = "Neutral",
  "4" = "Satisfied",
  "5" = "Very  Satisfied")
df = data.frame(
  year = sample(c(2023, 2022, 2020, 2018), 50, replace = TRUE),
  A = sample(likert_levels, 50, replace = TRUE),
  B = sample(likert_levels, 50, replace = TRUE),
  C = sample(likert_levels, 50, replace = TRUE))%>%
  mutate(across(everything(),as.factor))%>%
  as_tibble()%>%
  mutate(across(-year, ~ factor(.x, levels = likert_levels)))

# starting plot
likert <- df |>
  mutate(id = row_number()) |>
  pivot_longer(-c(id, year), names_to="group") |>
  pivot_wider(names_from=year) |>
  gglikert(c(`2023`, `2022`, `2020`, `2018`), 
           facet_rows=vars(group))

# get the order of years according to n
df2 <- df |>
  mutate(id = row_number()) |>
  pivot_longer(-c(id, year), names_to = "group") |>
  count(group, year) %>%
  arrange(n) %>%
  select(year) %>%
  distinct()

# change the order of years (.question) in the likert dataframe
likert$data <- likert$data |>
  mutate(.question = factor(.question, levels = df2$year))

# create the barplot
bar <- df |>
  mutate(id = row_number()) |>
  pivot_longer(-c(id, year), names_to = "group") |>
  count(group, year) |>
  ggplot(aes(x = n, y = fct_reorder(year, n))) +
  geom_col() +
  geom_label(aes(label = n),
             color = "white", hjust = 1,
             fill = NA, label.size = 0) +
  facet_wrap(~group, ncol = 1, strip.position = "right") +
  labs(y = NULL) +
  likert$theme

# plot them side by side
likert + bar + plot_layout(widths = c(0.8, 0.2), guides = "collect") &
  theme(legend.position = "bottom")

^{Created on 2024-12-13 with reprex v2.1.0}

If this isn't your desired outcome, please edit your question and explain exactly what you want your final plot to look like.