Environment Details

Please indicate the following details about the environment in which you found the bug:

• SDV version:
• Python version:
• Operating System:

Error Description

Currently, the create_parameters method will add invalid parameter values in cases like an entirely null column ('min_value' and 'max_value' both get set to None).

The create_parameters method should always return valid parameters - if a parameter can not be detected for a column for any reason, we should fall back to the expected default values.

Cases to check:

• Creating parameters for data with 0 rows
• Creating parameters for an entirely null column (both numerical and datetime)

Steps to reproduce

data = pd.DataFrame({'col': []})
metadata = Metadata().detect_from_dataframe(data)
DayZSynthesizer.create_parameters(data, metadata)