Make vec4f accept int parameter

[beaufortfrancois]

In the codelab for webgpu, the vertex module code is invalid and it will throw error when you try to use it. The reason behind this error is that the vec4f expects a float literal instead of a int literal and can't convert between them.

// wrong
@vertex
fn vertexMain(@location(0) pos: vec2f) ->
  @builtin(position) vec4f {
  return vec4f(pos.x, pos.y, 0, 1);
}

// correct
@vertex
fn vertexMain(@location(0) pos: vec2f) ->
  @builtin(position) vec4f {
  return vec4f(pos.x, pos.y, 0.0, 1.0);
}

My understanding according to https://gpuweb.github.io/gpuweb/wgsl/#f32-builtin is that it should work:

Note that it works fine with Tint, in Chrome browser.
See original issue raised at googlecodelabs/feedback#1305

[teoxoy]

Thanks for the report! This will be addressed by #4400.

[beaufortfrancois]

Thanks @teoxoy! Any rough ETA?

[teoxoy]

@jimblandy is actively working on it, since a lot of CTS shaders are running into this as well.