first crack at adapter-auto

[Rich-Harris]

#2805. For now this just adds adapter-vercel and adapter-netlify so that we can quickly try it out (difficult to test without publishing the package), but if it works we can immediately add other adapters (though I think for now it's really just Cloudflare Pages? not sure if it makes sense to try and detect that we should use adapter-node or adapter-static)

Please don't delete this checklist! Before submitting the PR, please make sure you do the following:

• It's really useful if your PR references an issue where it is discussed ahead of time. In many cases, features are absent for a reason. For large changes, please create an RFC: https://github.com/sveltejs/rfcs
• This message body should clearly illustrate what problems it solves.
• Ideally, include a test that fails without this PR but passes with it.

Tests

• Run the tests with pnpm test and lint the project with pnpm lint and pnpm check

Changesets

• If your PR makes a change that should be noted in one or more packages' changelogs, generate a changeset by running pnpx changeset and following the prompts. All changesets should be patch until SvelteKit 1.0

[changeset-bot]

🦋 Changeset detected

Latest commit: d5b7fe0

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 2 packages

Name	Type
@sveltejs/adapter-auto	Patch
create-svelte	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[Rich-Harris]

One alternative we might consider is keeping the test() logic but then only importing the underlying adapter if it's installed, otherwise prompting the user to install it. This eliminates the risk of adapter-auto growing crufty over time, but does introduce some additional friction.

An in-between option would be to install it automatically during the build (e.g. if process.env.VERCEL then we automatically install adapter-vercel). This adds some complexity though (need to consider different package managers etc), and probably isn't as fast as having it installed initially.

[Rich-Harris]

The adapter API could definitely use far more comprehensive documentation but a) probably best it doesn't derail this PR (since it's really just moving stuff from one file into another) and b) it should probably wait until the API is more final. i addressed the points above though