chore: implement bundle size action

[jasonsaayman]

Summary

Adds a Bundle Size GitHub Actions workflow that runs on PR open, synchronise, and reopen events, then comments bundle size comparisons for the v1 release stream.

Linked issue

N/A

Changes

• Add .github/workflows/bundle-size.yml with read-only contents permission and PR comment permission.
• Build axios with Node 24 using npm ci --ignore-scripts and npm run build.
• Run pinned axios/bundle-size against key dist/ artefacts and publish the comparison as a PR comment.

Checklist

• Tests added or updated (or N/A with reason: CI-only workflow change)
• Docs/types updated if public API changed (index.d.ts and index.d.cts) (N/A: no public API change)
• No breaking changes (or called out explicitly above)

---

Summary by cubic

Adds a GitHub Action that builds axios and reports bundle size changes on v1 PRs to catch regressions early. Comments only on non-fork PRs and writes a JSON report.

Description

• Add .github/workflows/bundle-size.yml with contents: read and pull-requests: write; runs on PR open/sync/reopen for the v1 stream.
• Build on Node 24 using npm ci --ignore-scripts and npm run build.
• Compare via axios/bundle-size@f15fb81ae6e3de06418f59498989e5aeeca9785d (release-stream: '1') for: dist/axios.js, dist/axios.min.js, dist/browser/axios.cjs, dist/node/axios.cjs.
• Output bundle-size-comparison.json and comment only when the PR is from this repo (not forks).

Docs

• Add a short /docs/ note about the “Bundle Size” workflow: when it runs, which files are compared, v1 stream scope, and that comments appear only on non-fork PRs.

Testing

• CI-only change; no source or tests modified. No additional tests needed.

Semantic version impact

• Patch (CI-only; no runtime or API changes).

^{Written for commit c3b5ed6. Summary will update on new commits. Review in cubic}

[github-actions]

Bundle Size Report

Compared current build against 1.x release stream baseline 1.16.1 for axios.

File	Baseline gzip	Current gzip	Difference	Status
dist/axios.js	41.0 KiB	41.0 KiB	0 B (0.00%)	🟢
dist/axios.min.js	20.0 KiB	20.0 KiB	0 B (0.00%)	🟢
dist/browser/axios.cjs	33.9 KiB	33.9 KiB	0 B (0.00%)	🟢
dist/node/axios.cjs	48.5 KiB	48.5 KiB	0 B (0.00%)	🟢
Total	143.4 KiB	143.4 KiB	0 B (0.00%)	🟢

Historical comparison: 1.x release stream baselines

Release	Baseline gzip	Current gzip	Difference	Status
1.16.1 latest	143.4 KiB	143.4 KiB	0 B (0.00%)	🟢
1.16.0	140.6 KiB	143.4 KiB	2.8 KiB (+1.99%)	🔵
1.15.2	128.2 KiB	143.4 KiB	15.2 KiB (+11.88%)	🔴
1.15.1	126.4 KiB	143.4 KiB	17.1 KiB (+13.49%)	🔴
1.15.0	123.1 KiB	143.4 KiB	20.3 KiB (+16.49%)	🔴
1.14.0	121.5 KiB	143.4 KiB	21.9 KiB (+18.00%)	🔴
1.13.6	122.4 KiB	143.4 KiB	21.0 KiB (+17.16%)	🔴
1.13.5	118.6 KiB	143.4 KiB	24.8 KiB (+20.89%)	🔴
1.13.4	120.8 KiB	143.4 KiB	22.7 KiB (+18.76%)	🔴
1.13.3	120.9 KiB	143.4 KiB	22.5 KiB (+18.60%)	🔴
1.13.2	119.7 KiB	143.4 KiB	23.7 KiB (+19.78%)	🔴

[cubic-dev-ai]

1 issue found across 1 file

Confidence score: 4/5

• This PR looks safe to merge overall, with only a minor CI workflow reliability concern rather than a product/runtime regression risk.
• In .github/workflows/bundle-size.yml, enabling comment-pr unconditionally can fail on fork-based pull_request runs because GITHUB_TOKEN is often read-only there, making PR comments non-writable.
• Pay close attention to .github/workflows/bundle-size.yml - gate comment-pr for non-fork PRs so comment steps don’t fail or behave inconsistently on forks.

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>