fix: ensure deferred effects can be rescheduled later on

[dummdidumm]

When deferring effects we didn't unmark the deriveds that lead to those effects. This means that they might not be reached in subsequent runs of mark_reactions.

Fixes #17118 (comment)

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
• Prefix your PR title with feat:, fix:, chore:, or docs:.
• This message body should clearly illustrate what problems it solves.
• Ideally, include a test that fails without this PR but passes with it.
• If this PR changes code within packages/svelte/src, add a changeset (npx changeset).

Tests and linting

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

[changeset-bot]

🦋 Changeset detected

Latest commit: 8f7e8b9

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

This PR includes changesets to release 1 package

Name	Type
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

[github-actions]

Playground

pnpm add https://pkg.pr.new/svelte@17147

[paoloricciuti]

Btw to fully solve this in SvelteKit I think we also need to discard the last fork when a new one come because I think right now we don't do it.

[dummdidumm]

Why though? It should be possible to have parallel forks. They could be independent and it would be very confusing to have a fork be discarded when another part of the app you may not have control over starts a fork

[paoloricciuti]

Mmm I'm specifically talking about preloading tho... shouldn't we discard older fork when you move your mouse on another link?

[dummdidumm]

Looking at SvelteKit's code it should alread do that (discard_load_cache)

[paoloricciuti]

It only do that inside goto, invalidateAll or if it errors...we also do that in preload before creating the new load_cache (this also causes await reactivity loss warnings). I might be wrong tho.

[dummdidumm]

Moving your mouse on another link should call preload and therefore discard the old cache, no?

[paoloricciuti]

Yeah but right now preload doesn't discard the cache unless it's an error...that's the issue.

[paoloricciuti]

Lul just realized a missed a "should" before the last sentence 😅

[PatrickG]

@paoloricciuti I think Rich already implemented it sveltejs/kit#14865 it is just not released yet.
But TBH I don't really understand why we need that. Wouldn't it be enough to discard all other forks when the user finally clicks a link.