Fix guide merging when there are repeated labels

[paleolimbot]

This is a PR to fix #3573, a bug whereby guides with repeated labels were merged incorrectly. This bug occurred because the two guides keys were merged using merge(), which only works when there are no duplicate values in the .label column.

Current behaviour:

guide1 <- list(
  key = tibble::tibble(
    colour = 1:4, 
    .label = c("low", "low", "high", "high")
  )
)

guide2 <- list(
  key = tibble::tibble(
    linetype = 1:4, 
    .label = c("low", "low", "high", "high")
  )
)

ggplot2:::guide_merge.legend(guide1, guide2)$key
#>   .label colour linetype
#> 1    low      1        1
#> 2    low      1        2
#> 3    low      2        1
#> 4    low      2        2
#> 5   high      3        3
#> 6   high      3        4
#> 7   high      4        3
#> 8   high      4        4

Behaviour after this PR:

guide1 <- list(
  key = tibble::tibble(
    colour = 1:4, 
    .label = c("low", "low", "high", "high")
  )
)

guide2 <- list(
  key = tibble::tibble(
    linetype = 1:4, 
    .label = c("low", "low", "high", "high")
  )
)

ggplot2:::guide_merge.legend(guide1, guide2)$key
#>   colour .label linetype
#> 1      1    low        1
#> 2      2    low        2
#> 3      3   high        3
#> 4      4   high        4

In the current implementation of merging, guides are only merged when they have identical direction, key$.label, title, and name. Because key$.label has to be identical for both guides, it should be safe to use cbind() in guide_merge.guide_legend() instead of merge(). I took the opportunity to write (non-visual) tests for guide merging to make sure it didn't break anything. As always, happy to be convinced otherwise (or that this shouldn't be fixed right now).

[clauswilke]

This looks good to me, but maybe get one more review before merging.

[thomasp85]

LGTM... I cannot predict whether we may get unexpected behaviour somewhere, but I guess time will tell 🙂

[lock]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/