Please describe the problem:
Rather than providing either HSV clockwise hue blends or HSV counterclockwise hue blends, gimp-edit-blend appears to choose a blend direction that avoids crossing the 0 degree hue line (red).  I expect to be able to duplicate the functionality for the UI.  That is, I expect to be able to request HSV clockwise hue blends or HSV counterclockwise hue blends via script-fu.

Steps to reproduce:
1. There are 4 scripts attached. Save them to your script folder.
2. Filters > Script-fu > Refresh Scripts.
3. Debug > HSV Gradient Test RGB 0 0 255 to 255 0 0.
4. Debug > HSV Gradient Test RGB 0 0 255 to 255 0 1.
5. Debug > HSV Gradient Test RGB 255 0 0 to 0 0 255.
6. Debug > HSV Gradient Test RGB 255 0 1 to 0 0 255.


Actual results:
3-4. Create new images. 3 goes from blue to red via green. 4 goes from blue to near-red via purple.
5-6. Create new images. 5 goes from red to blue via green. 6 goes from red to blue via purple.

Expected results:
3-4. Both images go from blue to red via the same color (green or purple, but not both).
5-6. Both images go from red to blue via the same color (purple or green, but not both, and the opposite color from steps 3-4).

Does this happen every time?
Yes

Other information:
The procedure browser currently lists the following blend-mode values for gimp-edit-blend:

The type of blend { FG-BG-RGB-MODE (0), FG-BG-HSV-MODE (1), FG-TRANSPARENT-MODE (2), CUSTOM-MODE (3) }

If we assume FG-BG-HSV-MODE (1) has been implemented as HSV with no 0 crossing, then there is a need for FG-BG-HSV-CW-MODE (4) and FG-BG-HSV-CCW-MODE (5).  If bug 141927 is implemented, there would also be a need for FG-BG-HSV-SHORT-MODE (6) and FG-BG-HSV-LONG-MODE (7).