Group specific prioritization setup

[ricschuster]

Adding some functionality in prioritizr to allow group specific prioritization, i.e. splitting up feature input by group.

library(dplyr)
library(tidyr)
df <- data.frame(id = seq(1:10),
                  group = round(2*runif(10),0),
                  feat1 = runif(10),
                  feat2 = runif(10))

df2 <- df %>%
     gather(feature_name, amount, -id, -group) %>%
     mutate(feature_name = paste0(feature_name, "_", group)) %>%
     spread(feature_name, amount, fill = 0)

group is the column that is used to identify groupings in the features.
The scenario is say a country wide setup (df), which is then processed
to also allow state specific priority setting (df2).