Add inversion checks

[ocefpaf]

After we compute density one can remove instabilities by checking for inversions and masking them. Most of the time those are due to CTD descending issue and/or not desired even when they are real features of the data (like when feeding the data to a numeric model.)

[ocefpaf]

Something like pmel_inversion_check:

    Additional clean-up and flagging of data after the SBE Processing.
    Look for inversions in the processed, binned via computing the centered
    square of the buoyancy frequency, N2, for each bin and linearly
    interpolating temperature, conductivity, and oxygen over those records
    where N2 <= -1 x 10-5 s-2, where there appear to be density inversions.

    NOTE: While these could be actual inversions in the CTD records, it is much
    more likely that shed wakes cause these anomalies.  Records that fail the
    density inversion criteria in the top 20 meters are retained, but flagged
    as questionable.