WGSL: Add binary + / - / * validation tests but for matrices

[Kangz]

Issue: #

---

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.

[jrprice]

We should also add some coverage for f32/f16 expressions whose results go out of range of the type.

Doesn't have to be in this CL, but just want to make sure we don't forget (e.g. add an .unimplemented() if you're not planning on doing it soon, or leave a comment on the spreadsheet about the remaining work).

[Kangz]

We should also add some coverage for f32/f16 expressions whose results go out of range of the type.

I'm slightly surprised that out-of-range concrete type expressions would be an error ... ah ok it's because WGSL has builtin consteval of everything? How do we want to test this though? It feels like it would require a lot of range-interval computations, more akin to what we do in execution tests.

[jrprice]

How do we want to test this though? It feels like it would require a lot of range-interval computations, more akin to what we do in execution tests.

Yes, we have helpers we can reuse (e.g. isRepresentable()). I'd take a look at the kind of thing we've done for some of the builtin validation tests, e.g. atan2. The values test generates ranges of values using helpers and then uses isRepresentable() to determine whether the result should be accepted or not.

We can't quite reuse validateConstOrOverrideBuiltinEval() for this, but we could make a similar helper for binary expressions.

As I say, don't have to do this in this PR if it turns into a bunch of extra work.

[Kangz]

Thanks for the explanation. I added an unimplemented but will relinquish further tasks as I'll be OOO for a bit.