Simplify LF/CLF API plumbing and reduce model/function duplication #12

the current API plumbing has too much duplication and redeclaration. Model names, wrappers, and function exposure are repeated in several places, which makes future extensions harder than necessary.

I would like to refactor the API so that ordinary luminosity functions and conditional luminosity functions are clearly separated, while allowing CLF functionality to reuse shared LF model machinery where appropriate. The API layer should stay thin and should not duplicate low-level model, integral, completeness, or redshift-density logic.

Main goals:

• separate LF and CLF user-facing APIs
• centralize model registration and metadata
• avoid redeclaring model names in multiple places
• expose bound-LF helper functions through shared wrappers
• make adding new LF/CLF models require minimal plumbing
• preserve existing numerical behavior

This is mainly a maintainability refactor so future models and extensions can be added cleanly and consistently.