How to speed up “shuf -n dt.csv” and setting column names using data.table?

Question

See the reproducible piece of code;

require(tidyverse); # To enable the use of %>%

# Generate and save data;
# With @N = 1000, and 6 variables
# and save loca