Update all dependencies + fix ensureRegistered bug

[thegedge]

I packed a bunch of stuff into one PR, but split it up into separate commits:

• Bumped most dependencies
  • Not bumping deep-freeze-es6, which jest is unhappy with for any version >= 2.
  • Not bumping eslint, because v9 isn't backwards compatible. Best to wait a few months
    for it to settle and the plugin ecosystem to support v9.
  • Fixed any linting issues due to bumping prettier/TS
• We were always checking the type argument passed to ensureRegistered for registration, and not each part of the prototype chain.
• Increases initCount in our benchmarks so that we don't get any unfortunate issues where a benchmark has some JITed and non-JITed samples. Not sure if the value is high enough to ensure things will be JIT-ed, but I figure it's still an improvement.

TODO

• Test out this branch in the Gadget repo (but opening for review early)
• Follow up with a separate PR to bump to 0.7.0

[codspeed-hq]

CodSpeed Performance Report

Merging #84 will not alter performance

_{Comparing bump-mqt (3b65711) with main (5f3020f)}

🎉 Hooray! codspeed-node just leveled up to 3.1.0!

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability 🥳!
Curious about what's new? Visit our releases page to delve into all the awesome details about this new version.

Summary

✅ 8 untouched benchmarks

[thegedge]

Oof, that's a rough regression. Lemme see if it's an easy fix, or if we don't bump.

[EDIT]
Not observing the same regression locally, except maybe on the "diverse root" benchmark (but it's not as bad as above)

[EDIT 2]
Looks like it could just be super high variance, which could be related to the JIT. I increased the amount of warmup runs to hopefully deal with such an issue. It may also be related to the bump in codspeed-node, which says that the latest version may potentially come with a regression.

[thegedge]

Debugging some issue with the fast instantiator:

Coming from InstantiatorBuilder.build

[thegedge]

☝️ rebasing to test the new benchmarks against mobx-state-tree directly, just to make sure any bumps to MST or mobx itself doesn't regress too much.

[thegedge]

I don't think those should have changed, so I'm going to amend the last commit and have it run again (edit: ✅)

[thegedge]

We had to patch out this change from the MST release in Gadget core, but that's because we do a ton of work with patches directly. All fixable, but that allows us to make progress more quickly. I think it's worth keeping these up to date though, so I'll leave the version bumps in!