perf: deduplicate CPU profile name conversions

[IlyasShabi]

Summary

This PR removes repeated V8 name conversions during eager CPU-profile translation.

Function names, script names, and script IDs are now retrieved once per profile node and reused for its line ticks and child nodes.
This avoids repeated UTF-8 decoding and V8 string-table lookups without adding a persistent cache.

Benchmark

A was added in benchmark/name-caching.js. Local measurements showed improvements of up to 17%, depending on the profile shape.

[github-actions]

Overall package size

Self size: 2.6 MB
Deduped: 3.31 MB
No deduping: 3.31 MB

Dependency sizes

| name | version | self size | total size | |------|---------|-----------|------------| | pprof-format | 2.3.1 | 504.33 kB | 504.33 kB | | source-map | 0.8.0 | 185.66 kB | 185.66 kB | | node-gyp-build | 4.8.4 | 13.86 kB | 13.86 kB |

_{🤖 This report was automatically generated by heaviest-objects-in-the-universe}

[szegedi]

Great little optimization! It must reduce memory usage as well, right, as there'll be fewer duplicate strings around? I presume V8 under the hood doesn't really do anything to avoid that.

[IlyasShabi]

@szegedi Thanks! I also reused the function name, script name, and script ID across child nodes and this increased the local improvement to up to 17% 🎉
On another hand for the heap profiler, V8 already deduplicates these strings, so this mainly saves repeated UTF-8 conversions and lookups.
I recently fixed a similar issue in V8's heap profiler: https://chromium-review.googlesource.com/c/v8/v8/+/8449311

[szegedi]

Nice!