Vendor rawler 0.7.2 unmodified

The crate as crates.io publishes it, minus .cargo-ok, its Cargo.lock and
data/testdata (13 MB of sample files only its own tests read). Not yet
routed through [patch.crates-io]; the next commit is the patch.

third_party/rawler-0.7.2/benches/perf.rs

@@ -0,0 +1,49 @@
+use criterion::{Criterion, criterion_group, criterion_main};
+use rawler::ljpeg92::LjpegCompressor;
+use std::hint::black_box;
+
+fn generate_data(w: usize, h: usize, ncomp: usize) -> Vec<u16> {
+  let mut img = vec![0; w * h * ncomp];
+
+  for (i, pix) in img.iter_mut().enumerate() {
+    *pix = i as u16 % 4u16;
+  }
+  img
+}
+
+fn encode_ljpeg(img: &[u16], w: usize, h: usize, ncomp: usize) {
+  let pred = 1;
+  let bps = 16;
+
+  let state = LjpegCompressor::new(img, w, h, ncomp, bps, pred, 0, 0).unwrap();
+  let _result = state.encode().unwrap();
+}
+
+fn copy_data(img: &[u16], w: usize, h: usize) {
+  let mut result = vec![0; w * h];
+  for (dst, src) in result.chunks_exact_mut(w).zip(img.chunks_exact(w)) {
+    dst.copy_from_slice(src);
+  }
+}
+
+fn criterion_benchmark(c: &mut Criterion) {
+  let mut group = c.benchmark_group("ljpeg-encoder");
+  // Configure Criterion.rs to detect smaller differences and increase sample size to improve
+  // precision and counteract the resulting noise.
+  group.significance_level(0.1).sample_size(20); //.measurement_time(Duration::from_secs(10));
+
+  let x = generate_data(3000, 2000, 1);
+
+  group.bench_with_input("encode_3000x2000", &x, |b, data| {
+    b.iter(|| encode_ljpeg(black_box(data), black_box(3000), black_box(2000), black_box(1)))
+  });
+
+  group.bench_with_input("copy_3000x2000", &x, |b, data| {
+    b.iter(|| copy_data(black_box(data), black_box(3000), black_box(2000)))
+  });
+
+  group.finish();
+}
+
+criterion_group!(benches, criterion_benchmark);
+criterion_main!(benches);