Add fetch to RequestEvent

[Rich-Harris]

Closes #6320. This moves SvelteKit's fetch wrapper to RequestEvent so that it can be used by server-only load functions and +server.js handlers as well as shared load functions. (It can even be used in handle, or — somewhat pointlessly — inside handleFetch, as a consequence of where it now lives.)

Marking as draft because we're no longer using (or able to use) GENERIC_ERROR, and I'm not quite sure what the consequences of that are just yet.

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 pnpm changeset and following the prompts. All changesets should be patch until SvelteKit 1.0

[changeset-bot]

🦋 Changeset detected

Latest commit: cc74f19

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

This PR includes changesets to release 1 package

Name	Type
@sveltejs/kit	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

[Conduitry]

Will the fetch that's passed to handleFetch now be identical to the event.fetch that's passed? It'd be a breaking change, but would this be a reason to stop passing fetch to keep things tidy?

[Rich-Harris]

Will the fetch that's passed to handleFetch now be identical to the event.fetch that's passed?

No. event.fetch invokes handleFetch({ event, request, fetch }); if the fetch argument was equivalent to event.fetch then it would call handleFetch({ event, request, fetch }), and you'd keep going round in circles.

[Conduitry]

I'm not sure where best to document this, but I do think we're probably overdue for a little bit of an explanation somewhere about how the different instances of fetch interact. I don't think it's enough to just add fetch as a method to the RequestEvent type. There are a number of nuances about browser vs server behavior, and about credential handling, and about calling handle instead of making a network request for same-origin fetches, etc. etc.

[Rich-Harris]

Agree, it could be clearer. Do you think that blocks this PR, or are we good to merge it in the meantime? #7087 is blocked on this one (though there is some discussion in that issue about rethinking docs as well)

[Conduitry]

@Rich-Harris Nope, I'm fine foregoing docs updates for now if we're going to be rethinking these docs in the context for #7087 anyway.