[deno] Rework error and device loss handling

[andyleiserson]

The goal of these changes is to make the CTS work again under deno, see #6838 and #7675.

There are two things happening here:

1. As was already noted by a TODO, reading device.lost should always return the same promise.
2. Handling errors in an async task without coordinating with the v8 runtime seemed to make things unhappy. I'm not sure exactly why -- the process was exiting silently. With this change, deno's V8TaskSpawner is used to schedule error handling through the runtime.

Testing
Tested using the webgpu:api,operation,device,lost:* and webgpu:api,operation,uncapturederror:* CTS tests. However, additional changes are needed to get the CTS to work, so until the rest of those changes are made, there isn't a mechanism to test this in the wgpu tree (that I'm aware of).

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.