Issue with mean_y when requesting B12/gnfw

[borisbolliet]

After updates of tabulation (thanks to Alina Sabyr) the large scale limit are incorrect, causing mean_y and dydz calculation to be incorrect for B12 and gnfw profiles. Other sz calculations are not affected.