VSX Version of vpx_sub_pixel_variance

[luctrudeau]

Implement a VSX version of :

• [ ] vpx_sub_pixel_variance8x8
• [ ] vpx_sub_pixel_variance16x16
• [ ] vpx_sub_pixel_variance32x32
• [ ] vpx_sub_pixel_variance64x64

Each function must:

• [ ] Pass the VpxSubpelVarianceTest suite
• [ ] Include Speed Test to the VpxSubpelVarianceTest suite (Disabled by default)
• [ ] Report performance in commit msg (compared to C version)
  • Must show significant speedup over C version

[luctrudeau]

Currently in libVPX, subpixel variance accounts for more than 15% of the run time for YouTube like encodes.

%	Function
5.59	vpx_sub_pixel_variance8x8_c
4.24	vpx_sub_pixel_variance16x16_c
3.25	vpx_sub_pixel_variance32x32_c
2.64	vpx_sub_pixel_variance64x64_c

There's a big potential here for speed ups.