Parameter processing for RLSSM models

**Current Implementation**
- I don't think there is a need to change the `from_user_specs()` function.
- But p_outlier handling might need a light wrapper around the likelihood function.

Start by tracing parameter processing for RLSSM models and reviewing from_user_specs(), then inspect the likelihood function where p_outlier is handled. The issue suggests a light wrapper may be needed, but it does not name files, tests, or a concrete completion criterion; clarify the intended behavior before making changes.

Written by the indexing model from the issue text.