benchmark: cut down http benchmark run time

[joyeecheung]

This split the http/simple.js benchmark so the header settings can be benchmarked differently to reduce the time spent on this benchmark. IIUC other parameters do not have an impact on the results, see the OP of #10558

Also cut the default duration to 5 seconds, on the benchmark CI we have about 20k req/sec for most types of scenarios so the results from 100k reqs should be representative enough.

cc @mscdex

Refs: #18003 (comment)

Checklist

• make -j4 test (UNIX), or vcbuild test (Windows) passes
• tests and/or benchmarks are included
• commit message follows commit guidelines

Affected core subsystem(s)

benchmark

[joyeecheung]

After #18373 lands we can see how the accuracy changes with the new parameters.

[joyeecheung]

CI: https://ci.nodejs.org/job/node-test-pull-request/12772/

[joyeecheung]

CI failures are unrelated

[joyeecheung]

Benchmark CI: https://ci.nodejs.org/job/benchmark-node-micro-benchmarks/107/

[joyeecheung]

For comparison, another one against a doc PR:
https://ci.nodejs.org/job/benchmark-node-micro-benchmarks/110/
https://ci.nodejs.org/job/benchmark-node-micro-benchmarks/111/

[BridgeAR]

Landed in 54b3a68

[MylesBorins]

Should this be backported to v8.x-staging or v6.x-staging? If yes please follow the guide and raise a backport PR, if not let me know or add the dont-land-on label.