Address bug in NORM operators

[PaulWessel]

The NORM operator in grdmath* and gmtmath says it will convert A so that max(A)-min(A) = 1. This it does, but not it a good way. For instance, if your data set generally goes from 3.0 to 3.5 and you call NORM, you get a result that goes from 6-7 instead - hardly a useful transformation. What is needed of course is to say that NORM converts A so that min(A) = 0 and max(A) = 1. This bug (and bad documentation) has now been fixed for both grdmath and gmtmath with this PR.

No tests were affected. Here is what it used to look like:

gmt math -T0/1/0.1 T 0.5 MUL 3 ADD =
0	3
0.1	3.05
0.2	3.1
0.3	3.15
0.4	3.2
0.5	3.25
0.6	3.3
0.7	3.35
0.8	3.4
0.9	3.45
1	3.5

gmt math -T0/1/0.1 T 0.5 MUL 3 ADD NORM =
0	6
0.1	6.1
0.2	6.2
0.3	6.3
0.4	6.4
0.5	6.5
0.6	6.6
0.7	6.7
0.8	6.8
0.9	6.9
1	7

and this is after the fix

gmt math -T0/1/0.1 T 0.5 MUL 3 ADD NORM =
0	0
0.1	0.1
0.2	0.2
0.3	0.3
0.4	0.4
0.5	0.5
0.6	0.6
0.7	0.7
0.8	0.8
0.9	0.9
1	1