try to fix d3_party() for new tidyr::nest() api

R/party.R

@@ -96,7 +96,7 @@ d3_party = function (tree=NULL, json=TRUE) {
  )
  colnames(counts) <- c("fitted", "response", "freq")
  counts$fitted = as.numeric(as.character(counts$fitted))
- counts <- tidyr::nest(counts, data=c(rlang::.data$response, rlang::.data$freq))
+ counts <- tidyr::nest(counts, .by = `fitted`)
  # would use dplyr join here, but nested data.frame
  # flattened on join; this does not happen with nested
  # tibble, but don't want to add another depedency