fix: run spinner onCancel on Ctrl+C

[cyphercodes]

What does this PR do?

Fixes spinner cancellation from Ctrl+C when the keypress is handled by the blocked input path. The spinner now treats the blocked-input exit path as cancellation, passes the configured input stream to block(), runs onCancel, and sets isCancelled.

Closes #573

Type of change

• Bug fix
• Feature
• Refactor (no behavior change)
• Documentation
• Performance improvement
• Tests
• Chore (dependencies, CI, tooling)

Checklist

• pnpm test passes (or targeted tests for my change)
• pnpm format has been run
• I have added/updated tests for my changes (if applicable)
• I have added a changeset

AI-generated code disclosure

• This PR includes AI-generated code

[github-actions]

Automation signals

@cyphercodes Activity patterns show signs of automation.

Classification: automation (score: 29)

Signal	Points	Detail
Multiple forks	+26	7 repositories forked in a single 24-hour window
Distributed PR spam pattern	+45	52 PRs spread across 51 different repositories (22.8 PRs/week)

_{Analyzed 200 public events via @unveil/identity}

[changeset-bot]

🦋 Changeset detected

Latest commit: fa6509f

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

This PR includes changesets to release 1 package

Name	Type
@clack/prompts	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

[gameroman]

These type assertions don't give me confidence

[43081j]

this isn't quite right. non-zero exits are not necessarily cancellations.

i think this needs discussion in an issue before a PR is made.

basically, we exit(0) by design when someone cancels:

clack/packages/core/src/utils/index.ts

Lines 54 to 58 in 8f1c380

	if (isActionKey([str, name, sequence], 'cancel')) {
	if (hideCursor) output.write(cursor.show);
	process.exit(0);
	return;
	}

so there's a discussion needed of if we want to continue doing that, or if we want to exit(1).