Gh 2809 faster hashcode collection equals

[hmottestad]

GitHub issue resolved: #

Briefly describe the changes proposed in this PR:

---

PR Author Checklist (see the contributor guidelines for more details):

• [ ] my pull request is self-contained
• [ ] I've added tests for the changes I made
• [ ] I've applied code formatting (you can use mvn process-resources to format from the command line)
• [ ] I've squashed my commits down to one or a few meaningful commits
• [ ] every commit message starts with the issue number (GH-xxxx) followed by a meaningful description of the change
• [ ] every commit has been signed off

[hmottestad]

Left is after, right is before.

Benchmark                                          Mode  Cnt     Score     Score   Error  Units
IsomorphicBenchmark.blankNodes                     avgt    2     5.570     4.998          ms/op
IsomorphicBenchmark.blankNodesLarge                avgt    2  7001.437  6964.143          ms/op
IsomorphicBenchmark.bsbm                           avgt    2    39.047    43.681          ms/op
IsomorphicBenchmark.bsbmArrayList                  avgt    2   135.787   141.421          ms/op
IsomorphicBenchmark.bsbmNotIsomorphic              avgt    2    26.620    26.989          ms/op
IsomorphicBenchmark.bsbmTree                       avgt    2    74.810    71.214          ms/op
IsomorphicBenchmark.empty                          avgt    2    ≈ 10⁻⁴    ≈ 10⁻⁴          ms/op
IsomorphicBenchmark.emptyNotIsomorphic             avgt    2    ≈ 10⁻⁵    ≈ 10⁻⁵          ms/op
IsomorphicBenchmark.internallyIsomorphic           avgt    2   343.667   330.034          ms/op
IsomorphicBenchmark.list                           avgt    2     7.555     7.533          ms/op
IsomorphicBenchmark.listEquals                     avgt    2     1.029     1.051          ms/op
IsomorphicBenchmark.longChain                      avgt    2     0.014     0.016          ms/op
IsomorphicBenchmark.manyProperties                 avgt    2    11.396    10.078          ms/op
IsomorphicBenchmark.manyProperties2                avgt    2     7.751     7.314          ms/op
IsomorphicBenchmark.sameModel                      avgt    2    ≈ 10⁻⁶    ≈ 10⁻⁶          ms/op
IsomorphicBenchmark.shacl                          avgt    2     3.166     3.068          ms/op
IsomorphicBenchmark.shaclValidationReport          avgt    2    10.082     9.506          ms/op
IsomorphicBenchmark.shaclValidationReport_changed  avgt    2     9.839     8.876          ms/op
IsomorphicBenchmark.sparqlTestCase                 avgt    2     0.054     0.051          ms/op
IsomorphicBenchmark.spinFullForwardchained         avgt    2    49.884    48.916          ms/op
IsomorphicBenchmark.uuid                           avgt    2     9.846    12.189          ms/op

[hmottestad]

At least on this very short benchmark run it didn't seem to make much difference.

[abrokenjester]

Still useful to know, thanks for trying this. I have some ideas around other ways to make some of this quicker, I just don't have the time right now to pursue them.

[hmottestad]

Yeah, thought it was cool. I read about using XOR just the other day on hackernews so wanted to give it a shot. Gonna leave this up here though and maybe I'll come back and run the benchmarks properly. That just takes so much time :(

[hmottestad]

I think we need some GC sensitive benchmarks for the isomorphic method.

[hmottestad]

Marked this as stale to signal that the PR hasn't been active for a while and that we should consider closing it.