perf: remove lto setting from CLI

[CrabNejonas]

What kind of change does this PR introduce?

• Bugfix
• Feature
• Docs
• New Binding issue #___
• Code style update
• Refactor
• Build-related changes
• Other, please describe:

Does this PR introduce a breaking change?

• Yes, and the changes were approved in issue #___
• No

Checklist

• When resolving issues, they are referenced in the PR's title (e.g fix: remove a typo, closes #___, #___)
• A change file is added if any packages will require a version bump due to this PR per the instructions in the readme.
• I have added a convincing reason for adding this feature, if necessary

Other information

This PR removes the lto=true setting for the CLI. This drastically improves the install performance (the time it takes to run cargo install tauri-cli).
Since the CLI is not doing performance-critical work and size is not important here since the cli will be compiled locally this does not have a downside it it afaik and only improves performance for rust users.

To give you an idea here are the install times on my MacBook Pro m2:

build	time
with lto	2:55.31
without lto	1:05.46

so as you can see disabling lto reduced the compile-time by 3x.

[FabianLars]

what do you think about moving the lto (and the new stuff you added in the js api) into a custom profile that inherits the release profile so that we can use that for the cargo-binstall CI?

[lucasfernog]

I had to use patch-package because napi-rs does not support custom profiles :(

[lucasfernog]

Also @JonasKruckenberg why did you target next here instead of dev?

[amrbashir]

I had to use patch-package because napi-rs does not support custom profiles :(

I would probably look into upstreaming this to napi-rs, the maintainer is pretty responsive and would probably publish it soon after merging.

Also if I understand correctly, this change wants to improve cargo install tauri-cli which iirc will use release profile by default which doesn't have this change and users would need to cargo install tauri-cli --profile release-size-optimized which is bad DX

[lucasfernog]

He's focusing on v3 so I'm not sure if he wants to patch for v2. I'll try anyway.

[lucasfernog]

I had to use patch-package because napi-rs does not support custom profiles :(

I would probably look into upstreaming this to napi-rs, the maintainer is pretty responsive and would probably publish it soon after merging.

Also if I understand correctly, this change wants to improve cargo install tauri-cli which iirc will use release profile by default which doesn't have this change and users would need to cargo install tauri-cli --profile release-size-optimized which is bad DX

I think you misunderstood. We don't want users to use release-size-optimized, we want them to have faster builds. The size optimization will be used only on the cargo-binstall and NPM assets.

[amrbashir]

ah right, I was even more confused by the name. LGTM then.

[lucasfernog]

@CrabNejonas we also need to rebase this branch with commit signing 😢

[socket-security]

👍 Dependency issues cleared. Learn more about Socket for GitHub ↗︎

This PR previously contained dependency changes with security issues that have been resolved, removed, or ignored.