refactor shapesys

Summary

with the custom_mod stuff we can quite cleanly refactor shapesys (and remove its finalize)

def required_parset(sample_data, modifier_data):
    # count the number of bins with nonzero, positive yields
    valid_bins = [
        (sample_bin > 0 and modifier_bin > 0)
        for sample_bin, modifier_bin in zip(modifier_data, sample_data)
    ]

    factors = [
        (nom_yield / unc) ** 2
        for is_valid, nom_yield, unc in zip(valid_bins, sample_data, modifier_data)
        if is_valid
    ]
    n_parameters = sum(valid_bins)
    return {
        "paramset_type": "constrained_by_poisson",
        "n_parameters": n_parameters,
        "is_shared": False,
        "is_scalar": False,
        "inits": (1.0,) * n_parameters,
        "bounds": ((1e-10, 10.0),) * n_parameters,
        "fixed": False,
        "auxdata": tuple(factors),
        "factors": tuple(factors),
    }

Additional Information

No response

Code of Conduct

• I agree to follow the Code of Conduct

[lukasheinrich]

yes this works

[lukasheinrich]

yes, this is a very quick / straight-forward PR