Make the CDEF RDO handle 4:2:2 properly

This fixes an assert:

av1/common/cdef_block.c:561: cdef_filter_fb: Assertion `bsize ==
BLOCK_8X8 || bsize == BLOCK_4X4' failed.

The RDO simply assigned a strength of 0 in the 4:2:2 case and called
cdef_filter_fb(), but cdef_filter_fb() will complain about 4:2:2 even
if the strength is 0.

The fix assigns a chroma mse of 0 when the the subsampling is
different for x and y rather than to call the filter.  This is faster
also.  The mse isn't really 0, but calculating the actual chroma mse
doesn't change result.

BUG=aomedia:881

Change-Id: I6154e21ddcca30e51baf805684dace10459c3350

av1/encoder/pickcdef.c

diff --git a/av1/encoder/pickcdef.c b/av1/encoder/pickcdef.c
index accc97e..cf6ccca 100644
--- a/av1/encoder/pickcdef.c
+++ b/av1/encoder/pickcdef.c
@@ -423,35 +423,40 @@
           int xsize = (nhb << mi_wide_l2[pli]) +
                       CDEF_HBORDER * (fbc != nhfb - 1) + xoff;
           sec_strength = gi % CDEF_SEC_STRENGTHS;
+          if (pli && !chroma_cdef) {
+            curr_mse = 0;
+          } else {
 #if CONFIG_CDEF_SINGLEPASS
-          copy_sb16_16(&in[(-yoff * CDEF_BSTRIDE - xoff)], CDEF_BSTRIDE,
-                       src[pli],
-                       (fbr * MI_SIZE_64X64 << mi_high_l2[pli]) - yoff,
-                       (fbc * MI_SIZE_64X64 << mi_wide_l2[pli]) - xoff,
-                       stride[pli], ysize, xsize);
-          cdef_filter_fb(NULL, tmp_dst, CDEF_BSTRIDE, in, xdec[pli], ydec[pli],
-                         dir, &dirinit, var, pli, dlist, cdef_count, threshold,
-                         sec_strength + (sec_strength == 3), pri_damping,
-                         sec_damping, coeff_shift);
-#else
-          if (sec_strength == 0)
             copy_sb16_16(&in[(-yoff * CDEF_BSTRIDE - xoff)], CDEF_BSTRIDE,
                          src[pli],
                          (fbr * MI_SIZE_64X64 << mi_high_l2[pli]) - yoff,
                          (fbc * MI_SIZE_64X64 << mi_wide_l2[pli]) - xoff,
                          stride[pli], ysize, xsize);
-          cdef_filter_fb(sec_strength ? NULL : (uint8_t *)in, CDEF_BSTRIDE,
-                         tmp_dst, in, xdec[pli], ydec[pli], dir, &dirinit, var,
-                         pli, dlist, cdef_count, threshold,
-                         sec_strength + (sec_strength == 3), sec_damping,
-                         pri_damping, coeff_shift, sec_strength != 0, 1);
+            cdef_filter_fb(NULL, tmp_dst, CDEF_BSTRIDE, in, xdec[pli],
+                           ydec[pli], dir, &dirinit, var, pli, dlist,
+                           cdef_count, threshold,
+                           sec_strength + (sec_strength == 3), pri_damping,
+                           sec_damping, coeff_shift);
+#else
+            if (sec_strength == 0)
+              copy_sb16_16(&in[(-yoff * CDEF_BSTRIDE - xoff)], CDEF_BSTRIDE,
+                           src[pli],
+                           (fbr * MI_SIZE_64X64 << mi_high_l2[pli]) - yoff,
+                           (fbc * MI_SIZE_64X64 << mi_wide_l2[pli]) - xoff,
+                           stride[pli], ysize, xsize);
+            cdef_filter_fb(sec_strength ? NULL : (uint8_t *)in, CDEF_BSTRIDE,
+                           tmp_dst, in, xdec[pli], ydec[pli], dir, &dirinit,
+                           var, pli, dlist, cdef_count, threshold,
+                           sec_strength + (sec_strength == 3), sec_damping,
+                           pri_damping, coeff_shift, sec_strength != 0, 1);
 #endif
-          curr_mse = compute_cdef_dist(
-              ref_coeff[pli] +
-                  (fbr * MI_SIZE_64X64 << mi_high_l2[pli]) * stride[pli] +
-                  (fbc * MI_SIZE_64X64 << mi_wide_l2[pli]),
-              stride[pli], tmp_dst, dlist, cdef_count, bsize[pli], coeff_shift,
-              pli);
+            curr_mse = compute_cdef_dist(
+                ref_coeff[pli] +
+                    (fbr * MI_SIZE_64X64 << mi_high_l2[pli]) * stride[pli] +
+                    (fbc * MI_SIZE_64X64 << mi_wide_l2[pli]),
+                stride[pli], tmp_dst, dlist, cdef_count, bsize[pli],
+                coeff_shift, pli);
+          }
           if (pli < 2)
             mse[pli][sb_count][gi] = curr_mse;
           else