Disable sampling of transformed variables from prior_predictive

[markvrma]

What are the (breaking) changes that this PR makes?

removed the if statement that appends rv_var.name to names and rv_var to vars_to_sample

resolves #5076

[ricardoV94]

Hi @markvrma thanks for opening the PR! These lines should also be updated:

pymc/pymc/sampling.py

Lines 2003 to 2006 in a823043

	var_names : Iterable[str]
	A list of names of variables for which to compute the posterior predictive
	samples. Defaults to both observed and unobserved RVs. Transformed values
	are not included unless explicitly defined in var_names.

And more importantly, SMC is relying on prior predictive to get a bunch of transformed samples here:

pymc/pymc/smc/smc.py

Lines 179 to 184 in e6fc2ec

	return sample_prior_predictive(
	self.draws,
	var_names=[v.name for v in self.model.unobserved_value_vars],
	model=self.model,
	return_inferencedata=False,
	)

That could will have to be refactored to make use of this function:

pymc/pymc/initial_point.py

Line 130 in d84c718

def make_initial_point_fn(

and calling it a bunch of times

[ricardoV94]

The lines above are also related to this logic

[ricardoV94]

Closing this PR due to inactivity. Feel free to reopen a new one anytime!