fix(material/checkbox): omit empty label element when no label is provided

[EduardF1]

What is the current behavior?

<mat-checkbox> always renders a <label> element inside mat-internal-form-field, even when no label content is projected. The empty <label> is hidden via display:none, but accessibility tooling such as Arc Toolkit still flags it as a violation. This is a regression vs. Material 14.

What is the new behavior?

The <label> element is now omitted from the DOM entirely when no label content has been projected. Detection happens in ngAfterContentInit by inspecting the projected light DOM children of the host element; the result drives an @if in the template via a _hasLabel signal.

The checkbox harness's _label locator is relaxed to locatorForOptional so getLabelText() returns '' when no label exists, instead of throwing.

Tests

Two assertions added to the existing 'without label' describe block in checkbox.spec.ts:

1. No <label> element is rendered when mat-checkbox has no projected content.
2. A <label> element with the expected text is rendered when content is projected.

I have not run the full Angular Components test suite locally (initial install + Bazel test would exceed my available time budget). CI should validate.

Does this PR introduce a breaking change?

No. The <label> is only removed when previously empty, and the [for]/click bubble paths gracefully handle its absence.

Fixes #33230

---

DCO: commit is signed off (Signed-off-by: Eduard Fischer-Szava <fischer_eduard@yahoo.com>).
CLA: I have read the CLA Document and I hereby sign the CLA — will respond to @googlebot if prompted.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[EduardF1]

@googlebot I have signed the CLA, please verify.

[crisbeto]

See #33230 (comment)

[EduardF1]

Thanks @crisbeto for the pointer to <mat-pseudo-checkbox/> - that's a cleaner fit for this case. Closing out here, appreciate the quick triage.