Validate data types of columns within incoming data.frames

[ThomUK]

As mentioned by @davidfoord1 in #123:

TODO:
Class check of indexed columns,
e.g., check_date(waiting_list[[referral_index]]).

Could be something to follow a more complete check_index().

[davidfoord1]

check_date() currently doesn't support any NA values as ambiguous dates. It will need a new mode to support different NA handling, so that users can still build their data frames with, for example, a logical column like removals <- c(NA, NA, NA)