propagating SED grid indices

[lea-hagen]

We need to clarify what grid indices (e.g., in the stats file) refer to which things and where. Here's what I've been able to come up with so far:

• specgrid_indx: Refers to the index in the spectral grid. In the SED grid, a given specgrid_indx value will appear multiple times because a given star has a range of dust properties applied to it.
• fullgrid_idx: When the SED grid is trimmed, this records the original index for each line. This column appears to be unused and is discarded at the fitting step.
• chi2min_indx: In the stats file, this is the SED grid index associated with the minimum chi^2. This is the index for whatever SED grid was used during the fitting (generally a trimmed one). When stats files are merged across source density, this column is retained, and a tag is added to specify the bin#_sub# that it goes with. When stats files are merged across subgrids, this keyword is discarded (though I suppose we could add another tag).
• Pmax_indx: I would expect this column to be identical to chi2min_indx, but it appears that only some of them are the same. I'm not sure why.

I'm wondering if we should more carefully propagate the grid index, so that info in the stats file can easily be used to grab data from either the SED or noisemodel files. Right now, if I want to look up the completeness associated with the best fit, I would use chi2min_indx, but that requires having the trimmed SED/noisemodel files available. It would be nice to trace it back to the index in the original untrimmed files.