Add the row_order() function

[rich-iannone]

This PR adds the row_order() function and it lets you modify the display order of rows within any row groups (or across the entire table). One or more column names can be provided to define the sorting criteria. Here's an example:

gtcars |>
  dplyr::select(mfr, model, ctry_origin, msrp) |>
  dplyr::filter(ctry_origin %in% c("Japan", "United Kingdom")) |>
  dplyr::slice_head(n = 12) |>
  gt(groupname_col = "ctry_origin") |>
  row_order(msrp, reverse = TRUE) |>
  fmt_currency(columns = msrp, decimals = 0) |>
  tab_options(column_labels.hidden = TRUE)

The ordering can also be constrained to named groups:

towny |>
  dplyr::filter(csd_type %in% c("city", "town", "township")) |>
  dplyr::select(name, csd_type, population_2021, density_2021) |>
  dplyr::slice_head(n = 5, by = csd_type) |>
  gt(groupname_col = "csd_type") |>
  fmt_integer(columns = population_2021) |>
  fmt_number(columns = density_2021, decimals = 1) |>
  row_order(density_2021, groups = c("city", "town"), reverse = TRUE)

Fixes: #1112