Unit test for life_exp() plot and gen_data_le() function

[SaranjeetKaur]

This PR adds unit tests for life_exp() and gen_data_le() functions.

Fixes #79 #89

Some housekeeping checks

• All ok when you run devtools::check() after the changes you have made?
• Have you updated the .Rd files using devtools::document()?
• If there were changes to README.Rmd, did you build and render it using devtools::build_readme()?
• If you added a new function or changed an existing function, have you created a (or updated an existing) unit test for it using usethis::use_test()?
• Everything looks ok?

[SaranjeetKaur]

@jzhu20 when computing gen_data_le() in data-process.R, it still uses the data_mean and looks for values like gender, age, etc.

Do you think this code needs to be modified? If yes, could you please suggest what needs to be changed?

gen_data_le <- function(data_mean) {
  data_le <- data_mean[, c(
    data_mean$source,
    data_mean$time,
    data_mean$gender,
    data_mean$age,
    data_mean$count,
    data_mean$deaths,
    data_mean$migrations
  )]
  data_le$timediff <- data_le$time - 2023
  # calculate the population considering both dead and alive
  # count is survivors, deaths = dead pop/count
  data_le$count_both <- data_le$count * data_le$deaths + data_le$count
  # calculate life expectancy px at each age
  data_le <- data_le[order(
    data_le$source,
    data_le$timediff,
    -data_le$age
  ), ]
  data_le <- data_le |>
    dplyr::group_by(
      data_le$source,
      data_le$timediff
    ) |>
    dplyr::mutate(
      tx = cumsum(data_le$count),
      px = data_le$tx / data_le$count
    )
  ## calculate period life expectancy as weighted sum of life expectancy at each age ##
  data_ple <- data_le |>
    dplyr::group_by(
      data_le$source,
      data_le$timediff
    ) |>
    dplyr::summarise(ple = stats::weighted.mean(data_le$px, data_le$count_both))

  ## calculate difference between baseline and intervention ##
  data_ple_wide <- tidyr::pivot_wider(data_ple,
    names_from = data_ple$source,
    id_cols = data_ple$timediff,
    values_from = c(data_ple$ple)
  )
  data_ple_wide$diff <- data_ple_wide$intervention - data_ple_wide$baseline

  return(data_ple_wide)
}