Default fitter in pintk should be determined using Fitter.auto()

[abhisrkckl]

Right now it is always GLSFitter.

[dlakaplan]

No, not true. It uses the same logic (but implemented separately) to select the fitter.
e.g.,

pintk tests/datafile/NGC6440E.par tests/datafile/NGC6440E.tim -v 
...
INFO     (pint.pintk.pulsar             ): Since wideband=False and correlated=False, selecting fitter=WLSFitter

[dlakaplan]

https://github.com/nanograv/PINT/blob/master/src/pint/pintk/pulsar.py#L107

[abhisrkckl]

I don't think it is the same logic. Fitter.auto() will choose DownhillWLSFitter instead of WLSFitter in this case.

(The always-GLSFitter thing was a mistake.)

[dlakaplan]

--downhill will choose the downhill variant. if needed that can be made the default.

[abhisrkckl]

--downhill will choose the downhill variant. if needed that can be made the default.

That's a better option, IMO, since the downhill fitters are supposed to be more robust.