fix: fatal failures hide the underlying error cause

[mrgrain]

When a deployment fails, the CLI now tells you what actually went wrong instead of just which stack it happened in.

Previously a fatal failure was flattened into a single line like ❌ Test-Stack-A failed: ResourceNotReady: <message>. That line was assembled by hand and thrown as a new error, discarding the original one — so anything nested below the first level was invisible. The AWS API error that ultimately caused the failure never made it to your terminal, and --debug didn't help either because the trace stopped at the first cause.

Failures now keep the original error attached as a cause, and the CLI prints the entire chain, one ‣ <name>: <message> line per level. You get the stack that failed, the CDK-level reason, and the underlying service error in one go, without re-running anything. With --debug you additionally get a stack trace for every level of the chain rather than only the outermost one.

Hotswap failures benefit the most from this. Errors coming out of a hotswap deployment are now labelled with the service responsible, so in a deployment touching several resources you can see which one broke instead of guessing. Cloud Control update failures previously surfaced as bare API errors with no hint of their origin; they now name the resource identifier and type that failed to update.

The two test apps switch from Stack.addDependency to addStackDependency — the former is deprecated in current aws-cdk-lib, and the deprecation output is annoying.

Before

After

Checklist

• This change contains a major version upgrade for a dependency and I confirm all breaking changes are addressed
  • Release notes for the new version:

---

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache-2.0 license

[github-actions]

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Scanned Files

None