feat(NODE-6539): add base napi C++ template with standard Node team tooling

[baileympearson]

Description

What is changing?

Adds a compilable and testable C++ addon that has the same API as the old compression implementation. Additionally, all shared tooling (lint, testing, and custom test tooling) is copied over.

A GHA is set up to run tests on ubuntu to demonstrate that the changes work as expected.

Is there new documentation needed for these changes?

What is the motivation for this change?

Double check the following

• Ran npm run format:js && npm run format:rs script
• Self-review completed using the steps outlined here
• PR title follows the correct format: type(NODE-xxxx)[!]: description
  • Example: feat(NODE-1234)!: rewriting everything in coffeescript
• Changes are covered by tests
• New TODOs have a related JIRA ticket

[nbbeeken]

I am surprised we're adding bindings here, don't we have a ticket to stop using this?

[baileympearson]

Yes. But I plan to align this package with our other native packages. When/if we get to the ticket you alluded to, we can address them all at once.

[nbbeeken]

I think it's unfortunate if we release with the known issues coming from that package. We can merge this, but do you think it's possible we can ship the final 2.0.0 without this? We can follow up

[baileympearson]

No, I don't think it is likely that we can complete that with all the other work we need to do in the next 8 work days (our timeline for this project). I am prioritizing consistency over a full overhaul/modernization of this package.

[nbbeeken]

My understanding is that we need to enumerate the possible file paths of the .node file so it doesn't seem like an overhaul. Meanwhile, we're potentially introducing a regression in our major release of this package since we didn't used to use this.

We likely could adapt the code in https://github.com/mongodb-js/zstd/blob/e7e8df676e3a6948b67708938495eb876d5bf117/bindings.js to just have a few different strings and it would do the work we need.

We can revisit trying to include this in the 2.0, doesn't need to hold up this PR.