feat!: Publish v1.0.0

[timfish]

This PR copies CODE_OF_CONDUCT.md, CONTRIBUTING.md, GOVERNANCE.md from import-in-the-middle to match how this project is managed. SECURITY.md still pointed to DataDog so it was deleted.

This PR is then marked feat!: so that a v1.0.0 will be published.

Many are using this in production and won't want breaking changes so I can't see any downsides to v1. This will mean that future feature additions do not require downstream users to keep updating their versions!

This PR should get agreement from all stakeholders.

[jsumners-nr]

Docs updates are good. But I am hesitant to say we are at a v1 yet.

[timfish]

Sorry, edited after your comment.

Many are using this in production and won't want breaking changes so I can't see any downsides to v1. This will mean that future feature additions do not require downstream users to keep updating their versions!

If we want breaking changes we can go to v2. If there anything in particular we need to do? Maybe remove the unused free() functions?

[jsumners-nr]

If we want breaking changes we can go to v2. If there anything in particular we need to do? Maybe remove the unused free() functions?

I think that question is exactly why we are not ready for v1. In the research I'm currently doing, I believe the free() function is going to come back in to play. So we need to establish the API surface contract for the object returned by the exported create function. And I think more experimentation is needed before we get to that point.

Maybe we remove it and then modify https://github.com/apm-js-collab/tracing-hooks/blob/527816fb60b31f1453fd73b39381da0caff28b06/index.js#L44 to invoke it if the function exists. But I doubt that's the only thing we still have to figure out.

In my view, the only reason to bump to v1 would be to get updates applied automatically through the semver contract. But I think we are still too early to accept updates so readily, and there are very few of us working with this at the moment. So, while I am very typically on the "don't use anything other than >=v1 modules" train, I think this is a case where we should take our time on that.

[timfish]

Fair points!

I'm going to close this and open another PR with these docs changes.

[timfish]

We're using both @apm-js-collab/code-transformer-bundler-plugins and @apm-js-collab/tracing-hooks so we have ensure they're both on the same version of @apm-js-collab/code-transformer so we don't end up with two different versions. A v1 would remove this anooyance!

There is no limit to the number or major versions we can release so I'm in the camp of sooner rather than later. We can release another major in a month or two. We're currently dealing with multiple releases per month that require updates and PRs and reviews, etc. Moving to v1 would reduce this burden.

[jsumners-nr]

I would open an issue and get feedback from all parties. I understand what you are saying, but I think that is to be expected as this is proved out.