See if the WavelengthCalibration1D.refine_fit penalty function can be made more robust

@kbwestfall noticed that the penalty function used to refine a wavelength calibration solution in `WavelengthCalibration1d` may lead to situations where the fit is improved by finding solutions that exclude lines rather than improving the polynomial. We need to take a look at how to make the penalty function more robust.

https://github.com/astropy/specreduce/pull/265#discussion_r2308206285

Start by reviewing the WavelengthCalibration1D.refine_fit implementation and the linked pull-request discussion about the penalty function. Determine how to assess cases where the fit excludes lines instead of improving the polynomial; done means the penalty is more robust against that behavior.

Written by the indexing model from the issue text.