Changes

Fixes for fill_holes and edsclr code (#1266)

* Fixing bugs in new hole fillers that caused error when ran in descending mode, the issue was that dz is negative in descending mode, so we need abs() around it when calculating the normalized mass. Also fixing a related bug in the way edsclr code calls the hole filler - it was not using the generalized bounds. Also removing the ifdef in favor of normal ifs

* Copying the rtm and thm values from arm_97_sounding.in to the edsclr_sounding file. This still allows arm_97 to catch the bugs we want it to.