arm: vp9itxfm: Reorder the idct coefficients for better pairing

arm: vp9itxfm: Reorder the idct coefficients for better pairing

All elements are used pairwise, except for the first one.
Previously, the 16th element was unused. Move the unused element
to the second slot, to make the later element pairs not split
across registers.

This simplifies loading only parts of the coefficients,
reducing the difference to the 16 bpp version.

This is cherrypicked from libav commit
de06bdfe6c8abd8266d5c6f5c68e4df0060b61fc.

Signed-off-by: Martin Storsjö <martin@martin.st>