Set featureLevel: "compatibility" when compatibility=1 is enabled

[beaufortfrancois]

As suggested in https://g-issues.chromium.org/issues/366151404#comment9, this PR adds a featureLevel CTS option.

Note that I'm not 100% sure I cover all changes required for the CTS.

---

Requirements for PR author:

• All missing test coverage is tracked with "TODO" or .unimplemented().
• New helpers are /** documented */ and new helper files are found in helper_index.txt.
• Test behaves as expected in a WebGPU implementation. (If not passing, explain above.)
• Test have be tested with compatibility mode validation enabled and behave as expected. (If not passing, explain above.)

Requirements for reviewer sign-off:

• Tests are properly located in the test tree.
• Test descriptions allow a reader to "read only the test plans and evaluate coverage completeness", and accurately reflect the test code.
• Tests provide complete coverage (including validation control cases). Missing coverage MUST be covered by TODOs.
• Helpers and types promote readability and maintainability.

When landing this PR, be sure to make any necessary issue status updates.

[kainino0x]

It's not immediately apparent that we will need to be able to run the CTS under customizable featureLevel values. It would be much simpler and avoid a lot of migration work if the CTS just keeps ?compatibility=1 as the option. We can add ?featureLevel=compatibility later if we find it's needed.

The thing that we need to do is have CTS request adapters using featureLevel: "compatibility" so that chromium can stop obeying compatibilityMode: true (but of CTS should keep compatibilityMode: true for now as well during migration).

[beaufortfrancois]

@kainino0x This makes this PR much simpler ;)

[greggman]

I think this should be changed so it doesn't request 'core'.

Instead of

    featureLevel: globalTestConfig.compatibility ? 'compatibility' : 'core',

It should be

    globalTestConfig.compatibility && { featureLevel: 'compatibility' },

wdyt?

[beaufortfrancois]

'core' is the default value for featureLevel so I think it's fine. See https://gpuweb.github.io/gpuweb/#dom-gpurequestadapteroptions-featurelevel

[greggman]

The problem is it makes the CTS never test that you get core if you don't request any feature level

[beaufortfrancois]

I believe this will be tested once featureLevel GPUAdapter attribute is added to the spec in gpuweb/gpuweb#5012?

[beaufortfrancois]

Note that I've added #4077 already