[CFL] Move store flag to CFL_CTX

With recent changes, it is now possible to store the storage
flag inside the CFL_CTX. This simplifies the implementation
and will allow reuse in the decoder.

This change does not alter the bitstream.

Change-Id: Ibb8aebdd3d06f8765d40248ece8a038892e87032
diff --git a/av1/encoder/rdopt.c b/av1/encoder/rdopt.c
index 695b4c4..7ea0319 100644
--- a/av1/encoder/rdopt.c
+++ b/av1/encoder/rdopt.c
@@ -2104,7 +2104,7 @@
                    OUTPUT_HAS_PREDICTED_PIXELS);
   }
 #if CONFIG_CFL
-  if (plane == AOM_PLANE_Y && x->cfl_store_y) {
+  if (plane == AOM_PLANE_Y && xd->cfl->store_y) {
     struct macroblockd_plane *const pd = &xd->plane[plane];
     const int dst_stride = pd->dst.stride;
     uint8_t *dst =
@@ -9839,15 +9839,15 @@
     // Don't store the luma value if no chroma is associated.
     // Don't worry, we will store this reconstructed luma in the following
     // encode dry-run the chroma plane will never know.
-    x->cfl_store_y = !x->skip_chroma_rd;
+    xd->cfl->store_y = !x->skip_chroma_rd;
 #else
-    x->cfl_store_y = 1;
+    xd->cfl->store_y = 1;
 #endif  // CONFIG_CB4X4
-    if (x->cfl_store_y) {
+    if (xd->cfl->store_y) {
       txfm_rd_in_plane(x, cpi, &this_rd_stats, INT64_MAX, AOM_PLANE_Y,
                        mbmi->sb_type, mbmi->tx_size,
                        cpi->sf.use_fast_coef_costing);
-      x->cfl_store_y = 0;
+      xd->cfl->store_y = 0;
     }
 #endif  // CONFIG_CFL
     max_uv_tx_size = uv_txsize_lookup[bsize][mbmi->tx_size][pd[1].subsampling_x]