refactor: Use napi-rs/image instead of sharp for screenshots

[Cadiac]

Summary

npx @n8n/computer-use ... commands failed silently with exit 1 on machines that have a system-wide libvips (e.g. installed via homebrew). Sharp's install script (useGlobalLibvips()in sharp/lib/libvips.js) detects a compatible system libvips and prefers to build from source against it - but building requires node-gyp, which isn't in sharp's deps. npx swallows the install error, so the user sees no banner, no logs, and no clue what went wrong.

I did consider adding node-gyp as a dependency but decided against it as that bundles a full C++ toolchain footprint for what amounts to one toJpeg helper.

@napi-rs/image (MIT) is a pure-Rust image library with statically linked prebuilds per platform. It has no system-library detection path and no build-from-source fallback, so install either succeeds with a prebuilt binary or fails loudly. Prebuild coverage is equivalent to sharp for our use case (desktop tool: macOS arm64/x64, Windows x64/arm64, Linux x64/arm64 glibc+musl). Admittedly I have no experience from using this library before, but napi-rs ecosystem seems reasonably actively developed.

Related Linear tickets, Github issues, and Community forum posts

Review / Merge checklist

• I have seen this code, I have run this code, and I take responsibility for this code.
• PR title and summary are descriptive. (conventions)
• Docs updated or follow-up ticket created.
• Tests included.
• PR Labeled with Backport to Beta, Backport to Stable, or Backport to v1 (if the PR is an urgent fix that needs to be backported)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[github-actions]

Performance Comparison

Comparing current → latest master → 14-day baseline

Idle baseline with Instance AI module loaded

Metric	Current	Latest Master	Baseline (avg)	vs Master	vs Baseline	Status
instance-ai-heap-used-baseline	186.75 MB	186.90 MB	186.41 MB (σ 0.28)	-0.1%	+0.2%	⚠️
instance-ai-rss-baseline	344.71 MB	345.77 MB	365.35 MB (σ 23.65)	-0.3%	-5.7%	✅

Memory consumption baseline with starter plan resources

Metric	Current	Latest Master	Baseline (avg)	vs Master	vs Baseline	Status
memory-heap-used-baseline	114.54 MB	114.54 MB	114.13 MB (σ 0.52)	+0.0%	+0.4%	✅
memory-rss-baseline	332.36 MB	332.36 MB	289.78 MB (σ 45.79)	+0.0%	+14.7%	✅

docker-stats

Metric	Current	Latest Master	Baseline (avg)	vs Master	vs Baseline	Status
docker-image-size-runners	386.00 MB	393.00 MB	391.78 MB (σ 10.35)	-1.8%	-1.5%	✅
docker-image-size-n8n	1269.76 MB	1269.76 MB	1269.76 MB (σ 0.00)	+0.0%	+0.0%	—

How to read this table

• Current: This PR's value (or latest master if PR perf tests haven't run)
• Latest Master: Most recent nightly master measurement
• Baseline: Rolling 14-day average from master
• vs Master: PR impact (current vs latest master)
• vs Baseline: Drift from baseline (current vs rolling avg)
• Status: ✅ within 1σ | ⚠️ 1-2σ | 🔴 >2σ regression

[Cadiac]

I'll test this a bit now to ensure it works, this didn't get enough testing last week.

[dlavrenuek]

Works perfectly, thank you @Cadiac 🚀

[Cadiac]

Also tested earlier locally that this did in fact did the issue I was seeing on my machine

[dlavrenuek]

Looks like I clicked on "comment" instead of "approve" last time 🙈

[n8n-assistant]

Got released with n8n@2.19.0