Update the Quark-ONNX quantization pass to support the latest mixed precision

[gengxinwu]

Describe your changes

This PR updates the Quark-ONNX quantization pass to support the latest mixed precision configuration APIs introduced in AMD Quark >= 0.11.0, including tensor-wise, op-type-wise, and layer-wise mixed precision quantization.

It also refactors configuration_preparation.py to use QLayerConfig.from_dict() and the new parse_data_type function instead of manually constructing individual data type specs (e.g., Int8Spec, UInt8Spec), making the code more maintainable and forward-compatible.

Checklist before requesting a review

• Add unit tests for this change.
• Make sure all tests can pass.
• Update documents if necessary.
• Lint and apply fixes to your code by running lintrunner -a
• Is this a user-facing change? If yes, give a description of this change to be included in the release notes.

(Optional) Issue link

[gengxinwu]

@jambayk Could you please help review this PR? Thanks!