feat: add radio component

[marjonlynch]

Add radio web component

• Chore: A change that does not impact distributed packages.
• Bug fix: A change that fixes an issue, link to the issue above.
• New feature: A change that adds functionality.

Is this a breaking change?

• This change causes current functionality to break.

Process & policy checklist

• I have added tests for my changes.
• I have tested my changes.
• I have updated the project documentation to reflect my changes.
• I have read the CONTRIBUTING documentation and followed the standards for this project.

[chrisdholt]

Similar to checkbox, I think our intent here is that this control behaves and is communicated to AT just as the native HTML radio buttons are input[type="radio"]. My primary concern is whether or not we have validated that is the case here. I've had feedback (linked in this PR as comments) on the W3C repo that label may not be valid for ARIA radios. Given that our goal here is parity with the platform, I think we should validate our assumption that this gives us parity from an AT standpoint. If we have parity with platform controls in terms of how these behave with AT, we're all good.

Two things to look for which are top of mind (not exhaustive):

1. Focus behavior differs from the platform for ARIA spec (noted here: https://w3c.github.io/aria-practices/#radiobutton)
2. Label behavior and validity may be different from HTML radios.

Again, the list is not exhaustive and my primary point is that we should validate this works as expected with assistive tech. My primary concerns here stem from feedback on existing implementations (react) where our aria radios do not have parity with the platform.

The last thing I will point out is that the W3C keyboard guidance calls out specific navigation behaviors for radios within a toolbar. What is our story for that? A possible solution: If we have parity with the platform in how these are communicated to AT, I would expect that implementation to match how you would do it with native platform controls. If we don't have parity with AT, I think we need to consider if we should map to the aria specification and the aria radio behaviors.