Switch heap profiling to use lazy allocation profile method by default

[IlyasShabi]

What does this PR do?:
Switch heap profiling to use mapAllocationProfile instead of the legacy getAllocationProfile, and remove the old code path.

Motivation:
PR #260 showed (via CI benchmarks) that mapAllocationProfile cuts memory usage by ~52% for a significant heap size. This change makes using mapAllocationProfile as the default for heap profiling.

How to test the change?:

• Unit tests

[github-actions]

Overall package size

Self size: 1.83 MB
Deduped: 2.2 MB
No deduping: 2.2 MB

Dependency sizes

| name | version | self size | total size | |------|---------|-----------|------------| | source-map | 0.7.6 | 185.63 kB | 185.63 kB | | pprof-format | 2.2.1 | 163.06 kB | 163.06 kB | | p-limit | 3.1.0 | 7.75 kB | 13.78 kB | | node-gyp-build | 3.9.0 | 8.81 kB | 8.81 kB |

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

[IlyasShabi]

@codex review

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 889cf467f1

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Avoid returning raw allocation root from v8Profile example

This example returns root out of v8Profile (pprof.heap.v8Profile(root => root)), but the allocation nodes are only valid during the callback; after it returns, the native allocation profile backing those node views is freed, so later property access can hit dangling data. Users who copy this snippet to keep a V8 profile object can get undefined behavior/crashes; the sample should instead transform/copy inside the callback and return that copied structure.

Useful? React with 👍 / 👎.

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-02-20 09:34:02

Comparing candidate commit b15bc7b in PR branch ishabi/use-map-allocation-profiler with baseline commit e88303f in branch main.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 93 metrics, 27 unstable metrics.