How do you customise spacing on the y-axis using ggplot2 in R?

Question

I want to create a forest plot of results from a heirarchical model so that the variable name appears on the y-axis and the names of each level in the heirarchy appear below (ideally indented from the variable name).

I have the following example:

library(tidyverse); theme_set(theme_classic())

# Create example data
data <- data.frame(variable = c(rep("Covariate1",3), rep("Covariate 2",3),
                                rep("Covariate3",3)),
                   posterior = rep(c("GroupA", "GroupB", "GroupC"),3),
                   mu = c(0.25, 0.08, 0.07, 0.27, -0.45, 0.01, 
                          0.37, 0.002, 0.24),
                   colour = rep(c("red","blue","yellow")))

data$est <- paste(data$variable, data$posterior, sep="_")

data
#     variable posterior     mu colour                est
#1  Covariate1    GroupA  0.250    red  Covariate1_GroupA
#2  Covariate1    GroupB  0.080   blue  Covariate1_GroupB
#3  Covariate1    GroupC  0.070 yellow  Covariate1_GroupC
#4 Covariate 2    GroupA  0.270    red Covariate 2_GroupA
#5 Covariate 2    GroupB -0.450   blue Covariate 2_GroupB
#6 Covariate 2    GroupC  0.010 yellow Covariate 2_GroupC
#7  Covariate3    GroupA  0.370    red  Covariate3_GroupA
#8  Covariate3    GroupB  0.002   blue  Covariate3_GroupB
#9  Covariate3    GroupC  0.240 yellow  Covariate3_GroupC


plot1 <- data |>
  ggplot(aes(y = fct_rev(reorder(est, as.numeric(row.names(data)))))) + 
  geom_point(aes(x=mu), shape=16, size=3.5, colour=data$colour) +
  geom_vline(xintercept = 0, linetype="solid", 
             linewidth=0.3, colour="gray70") +
  labs(x="Effect size", y="") +
  coord_cartesian(xlim=c(-0.5, 0.5), ylim=c(1,9)) +
  ggtitle("My Title")

plot1

I created the vector est to help explain the problem. I would like the y-axis labels to be more like this, using variable and posterior:

Covariate1
   GroupA
   GroupB
   GroupC

Covariate2
   GroupA
   GroupB
   GroupC

Covariate3
   GroupA
   GroupB
   GroupC

with a space between the covariates. Does anyone know how to do that? So far I've only found solutions for evenly increasing/decreasing space between all y-axis items, but I need something more customised to group my data.

I didn't use ggforestplot as I have a different statistical model for each covariate in the plot (plus confidence intervals, not included here) and I couldn't get it to work.