Add Jmh benchmark submodule

[spkrka]

Summary:
The distribution metric is faster when it's not contended,
but degrades during heavy contention.

Analysis:
Histogram throughput scales almost linearly with number of cores,
and maxes out at 52 ops/us, both with 4 and 8 threads.
Contention does not appear to be an issue there.

Distribution throughput is best when running with a single thread,
maxing out at 47 ops/us and drops down to 13 ops/us when running
with 4 threads. Going to 8 threads actually increases the throughput
though.

With a single thread, histogram runs in 0.055 us/op on average
and distribution runs in 0.021 us/op which is more than twice as fast.

However, when running with 4 threads concurrently, the distribution
average speed is 0.325 us/op (15x slower) compared to histogram which runs at
0.073 us/op (1.3x slower).

REMEMBER: The numbers below are just data. To gain reusable insights, you need to follow up on
why the numbers are the way they are. Use profilers (see -prof, -lprof), design factorial
experiments, perform baseline and negative tests that provide experimental control, make sure
the benchmarking environment is safe on JVM/OS/HW level, ask for reviews from the domain experts.
Do not assume the numbers tell you what you want them to tell.

Benchmark                                            Mode       Cnt      Score    Error   Units
DistributionBenchmark.dist1                         thrpt        10     47.352 ±  0.815  ops/us
DistributionBenchmark.dist2                         thrpt        10     16.576 ±  0.142  ops/us
DistributionBenchmark.dist4                         thrpt        10     13.032 ±  0.244  ops/us
DistributionBenchmark.dist8                         thrpt        10     26.443 ±  0.857  ops/us
DistributionBenchmark.hist1                         thrpt        10     18.029 ±  0.069  ops/us
DistributionBenchmark.hist2                         thrpt        10     35.253 ±  0.570  ops/us
DistributionBenchmark.hist4                         thrpt        10     52.909 ±  1.103  ops/us
DistributionBenchmark.hist8                         thrpt        10     53.726 ±  1.669  ops/us
DistributionBenchmark.dist1                          avgt        10      0.021 ±  0.001   us/op
DistributionBenchmark.dist2                          avgt        10      0.127 ±  0.007   us/op
DistributionBenchmark.dist4                          avgt        10      0.302 ±  0.012   us/op
DistributionBenchmark.dist8                          avgt        10      0.325 ±  0.030   us/op
DistributionBenchmark.hist1                          avgt        10      0.055 ±  0.001   us/op
DistributionBenchmark.hist2                          avgt        10      0.057 ±  0.001   us/op
DistributionBenchmark.hist4                          avgt        10      0.073 ±  0.003   us/op
DistributionBenchmark.hist8                          avgt        10      0.148 ±  0.006   us/op