Numpy 2.0 Support

We should update the code to be compatible with Numpy 2.0

This newer version has many deprecation that prevent using pyDVL with newer versions of numpy.
For example, `np.float_` was removed in the NumPy 2.0 release. We should use `np.float64` instead.

We can refer to its [migration guide](https://numpy.org/devdocs/numpy_2_0_migration_guide.html) for a complete list of changes.

Review the NumPy 2.0 migration guide linked in the issue to identify all deprecated usages like np.float_. Search the codebase for these patterns, likely across multiple files. Replace each deprecated item with its modern equivalent, then run the test suite to ensure compatibility.

由索引模型根据 Issue 内容生成。