Add transformer methods to PARSynthesizer

[amontanez24]

Problem Description

As a user, it would be useful to be able to update, learn or view transformers for my sequential data.

Acceptance criteria

Add the following methods to the PARSynthesizer

• auto_assign_transformers(data)
  • data is a pandas.DataFrame
  • By default, this method shouldn't assign any transformers.
• get_transformers()
  • This should return the transformers for every column in the data.
• update_transformers(column_name_to_transformer)
  • column_name_to_transformer is a dict mapping column names to transformer instances
  • If the user tries to update the transformer for a context column, we should raise an error

Expected behavior

>>> synthesizer.assign_transformers(data)

>>> synthesizer.get_transformers(table_name='users')
{
  'patient_id': None
  'ssn': None,
  'time': None
  ...
}

>>> from rdt.transformers as rt

>>> paf = rt.PeudoAnonymizedFaker(provider_name='address', function_name='address')
>>> pn = rt.AnonymizedGeoExtractor()
>>> pn2 = rt.AnonymizedGeoExtractor()

>>> model.update_transformers(
  column_name_to_transformer={
    'patient_id': rt.RegexGenerator(regex_format='[0-9]{4}'),
})

Error: Transformers for context columns are not allowed to be updated.

Additional context

• We should not return or allow the modification of any of the transformers in the context synthesizer. For now, that will be hidden from the user.