unary

This release introduces DFlash2 support with local convolution and candidate selection, alongside various bug fixes and code refactoring. Key changes include optimizations for DFlash2 and fixes for mrope and graph number calculations.

This release addresses several failing unit tests related to core operations like ACC, L2_NORM, and fused_glu. It also provides updated pre-built binaries for a wide range of operating systems and hardware configurations.