P92 fitting tests failing

[karllark]

Need to investigate. Details in PR #186

[icweaver]

Hi Karl, thanks for this awesome package!

While working on adding P92 to our JuliaAstro implementation, we noticed that you seemed to change the default parameter value for FUV_lambda from the paper value of 0.08 μm in Table 4 of Pei, to a value of 0.07 μm.

Could this be related to the fitting issues that you are seeing here? Apologies if this is unrelated, we just wanted to flag this in case there was some other reasoning for using this new value.

[karllark]

Interesting. Will investigate.

[karllark]

Investigated. Sadly, not the origin of the failure. Continuing to investigate. It may just be that the best fit parameters are just a bit different due to upstream changes in code that affects the fitting. Will investigate more.