hal/vulkan: Fix multi draw indirect emulating and respect maxDrawIndirectCount

[beicause]

Description
Noted this when I was checking #8162

The offset argument of draw_indexed_indirect in hal/vulkan seems incorrect. It should + i * stride.
Also multi draw indirect should respect limits.maxDrawIndirectCount.

Actually, vkCmdDrawIndirectCount vkCmdDrawIndexedIndirectCount vkCmdDrawMeshTasksIndirectCountEXT also require that the count in the countBuffer must be <= maxDrawIndirectCount, altough on many devices it's u32::MAX(vulkan gpuinfo)

Testing
I assume the code is correct, I don't have a device to test this.

Squash or Rebase?
Squash

Checklist

• Run cargo fmt.
• Run taplo format.
• Run cargo clippy --tests. If applicable, add:
  • --target wasm32-unknown-unknown
• Run cargo xtask test to run tests.
• If this contains user-facing changes, add a CHANGELOG.md entry.

[ErichDonGubler]

I'm not @cwfitzgerald (the current assignee), but: We definitely need test coverage for this. This is fundamental enough to these specific APIs that I'd be shocked if there wasn't a CTS test path we could simply drop into cts_runner/test.lst it's necessary.

EDIT: This isn't covered by CTS, because it ain't WebGPU. 😅

[cwfitzgerald]

if there wasn't a CTS test path

MDI currently isn't in the spec, though there are plans to add it at some point.

We definitely need test coverage for this

+1

[ErichDonGubler]

Welp, I'm shocked, and it's my fault. 🤦🏻‍♂️😆

[cwfitzgerald]

Looks good, lets get some more tests and one quibble

[beicause]

Not sure how I should add tests for this as I’m not very familiar with this repository.

[cwfitzgerald]

Information on the tests are available in https://github.com/gfx-rs/wgpu/blob/trunk/docs/testing.md if you're interested but honestly we can keep his as is for now, landing this fix is more important than getting testing for it. Lets just get that one change implemented and we can land this.

Stale