Update FFI to be more portable and direct.

[natefaubion]

Description of the change

This updates the timing FFI to not bind to hrtime directly, which is an odd API in general, instead taking the closure directly and returning the expected nanoseconds.

My primary motivation for doing this is that the existing framework does not work well with purescript-backend-optimizer, which aggressively eliminates unused code. The current pattern after inlining results in an unused call to f, which means nothing ever gets executed. By using an opaque FFI call, the demand on the closure is kept, and everything works as expected. Normally I wouldn't suggest doing this arbitrarily for this tool, but I think this is a better binding anyway, and is likely easier for other backends to implement since it doesn't rely on the specifics of a Node API.

---

Checklist:

• Added the change to the changelog's "Unreleased" section with a reference to this PR (e.g. "- Made a change (#0000)")
• Linked any existing issues or proposals that this pull request should close
• Updated or added relevant documentation
• Added a test for the contribution (if applicable)

[natefaubion]

I'm not sure what core's policy is on module private FFI implementations. Do we consider this a breaking change?

[JordanMartinez]

I don't think we have a clear policy on whether or not changing FFI counts as a breaking change. I think this is fine though.

[JordanMartinez]

@thomashoneyman / @garyb Thoughts on this?

[garyb]

LGTM 👍 I'd be happy with this even if we were just doing it to make the interface better.

Seems I wasn't watching the library, so just got the notification with the ping.

[JordanMartinez]

@natefaubion Mind adding a changelog entry?

[thomashoneyman]

This is internal so I don’t consider it breaking.

[natefaubion]

I've updated the changelog.

[JordanMartinez]

Can this get a release?