feat: GPU-accelerated SVD and quaternion conversion for ~4x speedup

[Mozoloa]

This PR adds optional GPU acceleration to the covariance matrix decomposition and rotation-to-quaternion conversion, providing significant performance improvements for CUDA users while maintaining backward compatibility.

Changes

linalg.py

• Add use_gpu parameter to quaternions_from_rotation_matrices() (default: True)
• Add pure PyTorch GPU implementation using Shepperd's method
• Original scipy CPU implementation preserved as fallback
• ~300x faster for large batches (2M+ gaussians)

gaussians.py

• Add use_gpu parameter to decompose_covariance_matrices() (default: True)
• GPU path: SVD on GPU + vectorized reflection correction
• CPU path: original float64 behavior preserved for maximum precision
• Automatic fallback to CPU if GPU SVD fails

Performance

Tested on RTX 4090 with ~700k gaussians per frame:

• Before: ~4.0s per frame (3s quaternion conversion on CPU)
• After: ~1.0s per frame
• 4x overall speedup

The bottleneck was scipy.spatial.transform.Rotation.from_matrix() which requires CPU transfer and numpy conversion. The new GPU implementation stays entirely on device.

Backward Compatibility

• Default behavior unchanged for CPU tensors
• Set use_gpu=False to force original CPU behavior
• API is fully backward compatible (new parameter has default value)