Minor updates to mitscherlich( ) and respective vignette, and built without errors

As mentioned in commit,

place input variable as first argument in NLS equations
add error message if data is not on percentage scale
move error messages earlier to catch errors earlier and prevent unnecessary steps
equation now has negative (-) sign for c
equation in table output also updated
CSTV calculation adjusted for that (results verified)
updated estimation for starting C value
updated upper and lower limits as previous limits were caused an issue on dataset tested

adriancorrendo / soiltestcorr