Feature Request：Add some parameters to `geom_smooth` to adjust the properties of the lines at the edges of the confidence intervals.

Hi, Teun!
In the process of using ggplot for plotting fitted lines, I came across a scenario of trying to show confidence intervals using line segments rather than shaded space. I asked a question on stackoverflow, But I don't think it's a very elegant or user-friendly solution.

https://stackoverflow.com/questions/79668693/plotting-the-95-error-margin-segments-of-the-fitted-line-and-controlling-the-li

I used the following code approach to achieve the results I wanted to show:

library(ggplot2)

ggplot(data = mpg, aes(x = displ,y = hwy))+
  geom_point()+
  geom_smooth(method = "lm", se = FALSE, color = "black") +
  stat_smooth(method = "lm", 
              geom = 'ribbon',
              fill = NA,
              linetype = 2,
              se = TRUE, color = "blue")
#> `geom_smooth()` using formula = 'y ~ x'
#> `geom_smooth()` using formula = 'y ~ x'

^{Created on 2025-07-09 with reprex v2.1.1}

I know that the code actually does the same calculation twice for the same process.

Is there any chance that similar to the previous additions to the whiskers, staples and median line related properties of geom_boxplot, geom_smooth could be adjusted appropriately as well? It's not even about aesthetics, it's about simply adjusting basic properties like colour/color, linewidth and linetype.

I hope I'm not disturbing you and interfering with the release of a new version of ggplot.

Best wishes,
Hu

Start with the geom_smooth() and stat_smooth() entry points shown in the issue, and reproduce the supplied mpg example to understand the desired confidence-interval edge lines. Determine how color, linewidth, and linetype should apply to those edges without requiring the duplicated smoothing calls; done means the requested properties can be adjusted through the smooth-layer interface.