Sweep `pacta.data.preparation/tests/testthat/helper-fake_data.R` for data faking functions

[jdhoffa]

And replace those functions with consolidated and centralized faking functions.

nit: maybe add a fixme/comment about eventually being able to replace this with functions from pacta.data.validation? I'd rather see these faker functions consolidated into one location rather than duplicated many places (possibly with slight variation)

Originally posted by @cjyetman in https://github.com/RMI-PACTA/pacta.data.preparation/pull/286#discussion_r1159527963

https://github.com/RMI-PACTA/pacta.data.preparation/blob/main/tests/testthat/helper-fake_data.R

AB#10853

[jdhoffa]

To some extent this is related to #32, however I see it as subtly different.

On the one hand, it is useful to have a representative dataset to use for regression testing (a dataset that contains all different permutations of expected input) that would be #32

On the other hand, it's useful to have a minimal (one-row) dataset, that can be used and iterated on quickly for the purposes of unit testing. That is this issue.