Is the `race = FALSE` flag working correctly?

[AdrienLeGuillou]

when calling:

  epistats <- build_epistats(
    geog.lvl = "city",
    geog.cat = "Atlanta",
    init.hiv.prev = 0.33,
    race = FALSE,
    time.unit = time_unit
  )

we get an error because init.hiv.prev is of length 1 and not 3.
Is this the expected behavior?
Then looking quickly into the object I still see the 3 races.

I need to look into it more. @smjenness am I missing something obvious here before I go dive deeper?

Note: this behavior was present before Pr #57