`chop()` returns the wrong result with empty data frames

[DavisVaughan]

library(tidyr)

df <- tibble(x = 1, y = 2)

chop(df, y)
#> # A tibble: 1 × 2
#>       x           y
#>   <dbl> <list<dbl>>
#> 1     1         [1]

# Should be y=list<dbl>
chop(df[0,], y)
#> # A tibble: 0 × 2
#> # … with 2 variables: x <dbl>, y <dbl>

^{Created on 2021-11-09 by the reprex package (v2.0.1)}