add a regression test for the fractional canvas size

[HuzaifaChaudary]

while looking at #12269 i went to find why 4.5.1 behaves differently from 4.5.0 on a responsive chart and it comes down to retinaScale.

in 4.5.1 the canvas dimensions are written straight from round1:

const deviceHeight = round1(chart.height * pixelRatio);
const deviceWidth = round1(chart.width * pixelRatio);
...
if (chart.currentDevicePixelRatio !== pixelRatio
    || canvas.height !== deviceHeight
    || canvas.width !== deviceWidth) {
  canvas.height = deviceHeight;
  canvas.width = deviceWidth;

canvas.width and canvas.height are unsigned long so the browser truncates whatever is assigned to them. when the container measures to a fraction the stored value can never equal the value it is compared against. the check is then true on every call and retinaScale keeps reporting that the size changed.

i measured it in chrome headless:

canvas.width = 600.8            -> 600
4.5.1 logic on 300.4 x 150.2    -> true true true
master on the same size         -> true false false

this matters because _resize uses the return value as its early exit:

if (!retinaScale(this, newRatio, true)) {
  return;
}

so on 4.5.1 a fractional size makes every attach and every resize fall through to _doResize instead of stopping there. that is the path in the stack trace on #12269.

#12142 already fixed this by flooring the canvas dimensions but it landed without a test and it is not in a release yet. this adds the missing guard so the behaviour does not slip back. putting the floor back out makes the new test fail with Expected true to be false on the second call.

no source change here.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.