Improve performance in searchstream

[Uzlopak]

Unit tests are green, so I think the change provides an equivalent functionality.

Benchmark:

> @fastify/[email protected] bench:dicer
> node deps/dicer/bench/dicer-bench-multipart-parser.js

9090.91 mb/sec
11111.11 mb/sec
10000.00 mb/sec
11111.11 mb/sec
11111.11 mb/sec
11111.11 mb/sec
12500.00 mb/sec
12500.00 mb/sec
11111.11 mb/sec
11111.11 mb/sec

Checklist

• run npm run test and npm run benchmark
• tests and/or benchmarks are included
• documentation is changed or added
• commit message and code follows the Developer's Certification of Origin
  and the Code of conduct

[Uzlopak]

Also deoptimization traces are gone.

node --trace-deopt --trace-opt ./deps/dicer/bench/dicer-bench-multipart-parser.js

[Uzlopak]

I think this PR is actually good enough performance gain for this iteration, so that even @mcollina would be happy :-D as it does not only shaves some yoctoseconds.

So we improved the performance from about 3,5 GB/s to 10-12 GB/s.

[kibertoad]

will take a look later tonight, need to wrap up version release at work

[mcollina]

lgtm

[kibertoad]

@Uzlopak BTW, I've published a beta version of the package and tried running fastify-multipart as well as multer v2 tests against it, and all of them are still passing, so looks like your optimizations didn't break anything :)

For the first public release I still want to wait for #26
Is there anything else that is on your "must have" list for 1.0.0?

[Uzlopak]

@kibertoad
I think we already achieved alot :). Just let us merge the last PRs and then release it :)