Optimise sad_plane_internal

This is part of a series of commits authored by @maj160 to improve performance of rav1e.

This commit results in an overall performance improvement of about 1% at speed 2 10-bit:

```
Benchmark 1: ~/Downloads/rav1e_master -s 2 --quantizer 64 ~/xiph-media-files/objective-1-fast-10bit/speed_bag_640x360_60f.y4m -y -o /dev/null
  Time (mean ± σ):     43.492 s ±  0.088 s    [User: 43.514 s, System: 0.167 s]
  Range (min … max):   43.329 s … 43.603 s    10 runs

Benchmark 2: ~/Downloads/rav1e_mod -s 2 --quantizer 64 ~/xiph-media-files/objective-1-fast-10bit/speed_bag_640x360_60f.y4m -y -o /dev/null
  Time (mean ± σ):     43.062 s ±  0.086 s    [User: 43.068 s, System: 0.173 s]
  Range (min … max):   42.947 s … 43.224 s    10 runs
```

Author: maj160

Cargo.lock

@@ -88,7 +88,7 @@ dav1d-sys = { version = "0.7.0", optional = true }
 aom-sys = { version = "0.3.2", optional = true }
 scan_fmt = { version = "0.2.3", optional = true, default-features = false }
 ivf = { version = "0.1", path = "ivf/", optional = true }
-v_frame = "0.3.0"
+v_frame = "0.3.3"
 av-metrics = { version = "0.9.0", optional = true, default-features = false }
 rayon = { package = "maybe-rayon", version = "0.1", default-features = false }
 crossbeam = { version = "0.8", optional = true }

Cargo.toml

@@ -29,22 +29,16 @@ pub(crate) mod rust {
     src: &Plane<T>, dst: &Plane<T>, _cpu: CpuFeatureLevel,
   ) -> u64 {
     debug_assert!(src.cfg.width == dst.cfg.width);
-
-    let width = src.cfg.width;
+    debug_assert!(src.cfg.height == dst.cfg.height);
 
     src
       .rows_iter()
       .zip(dst.rows_iter())
       .map(|(src, dst)| {
-        let src = src.get(..width).unwrap_or(src);
-        let dst = dst.get(..width).unwrap_or(dst);
-
         src
           .iter()
           .zip(dst.iter())
-          .map(|(&p1, &p2)| {
-            (i16::cast_from(p1) - i16::cast_from(p2)).unsigned_abs() as u32
-          })
+          .map(|(&p1, &p2)| i32::cast_from(p1).abs_diff(i32::cast_from(p2)))
           .sum::<u32>() as u64
       })
       .sum()