Fixed some performance issues in DTW.

[zond]

No description provided.

[zond]

Before:

zond@nyarlathotep:~/projects/zimtohrli/build{main}$ ninja && ./zimtohrli_benchmark "--benchmark_filter=.*DTW.*"
[7/7] Generating /usr/local/google/home/zond/projects/zimtohrli/go/goohrli/goohrli.a
2024-06-13T15:12:56+00:00
Running ./zimtohrli_benchmark
Run on (128 X 2450 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 4.10, 5.45, 5.21
---------------------------------------------------------------------------
Benchmark                 Time             CPU   Iterations UserCounters...
---------------------------------------------------------------------------
BM_DTW/100         23418941 ns     23418196 ns           30 items_per_second=4.27018k/s
BM_DTW/512        616775362 ns    616719164 ns            1 items_per_second=830.2/s
BM_DTW/1000      2376034758 ns   2375986230 ns            1 items_per_second=420.878/s
BM_ChainDTW/100    23517395 ns     23515845 ns           30 items_per_second=4.25245k/s
BM_ChainDTW/512   320192858 ns    320187047 ns            2 items_per_second=1.59907k/s
BM_ChainDTW/4096 3703533277 ns   3703360514 ns            1 items_per_second=1.10602k/s
BM_ChainDTW/5000 4565519120 ns   4565471480 ns            1 items_per_second=1.09518k/s

After:

zond@nyarlathotep:~/projects/zimtohrli/build{fix-dtw}$ ninja && ./zimtohrli_benchmark "--benchmark_filter=.*DTW.*"
[7/7] Generating /usr/local/google/home/zond/projects/zimtohrli/go/goohrli/goohrli.a
2024-06-13T15:13:40+00:00
Running ./zimtohrli_benchmark
Run on (128 X 2450 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 8.54, 6.25, 5.48
---------------------------------------------------------------------------
Benchmark                 Time             CPU   Iterations UserCounters...
---------------------------------------------------------------------------
BM_DTW/100          1980830 ns      1980811 ns          351 items_per_second=50.4844k/s
BM_DTW/512         53553099 ns     53552272 ns           13 items_per_second=9.56075k/s
BM_DTW/1000       203519392 ns    203519992 ns            3 items_per_second=4.91352k/s
BM_ChainDTW/100     1967938 ns      1967915 ns          355 items_per_second=50.8152k/s
BM_ChainDTW/512    26881296 ns     26880065 ns           26 items_per_second=19.0476k/s
BM_ChainDTW/4096  327046297 ns    327026156 ns            2 items_per_second=12.525k/s
BM_ChainDTW/5000  390924859 ns    390919898 ns            2 items_per_second=12.7903k/s