Fix ARM NEON wide-gamut color conversion

[gregbenz]

What changed

• Use BT.2100 coefficients for BT.2100 RGBA8888 input.
• Use BT.601 luma coefficients for Display-P3, matching the existing scalar conversion.
• Add an ARM-only regression test comparing NEON and scalar output for saturated colors.

Why

The ARM NEON path used the wrong color matrices for Display-P3 and BT.2100 input. This could cause visibly different encoded color on ARM devices compared with other platforms.

Testing

• The new regression fails on current main and passes with this fix on Apple Silicon.
• Full non-HEIF unit suite: 1,067 passed, 224 skipped, 0 failed.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.