Update Windows xformers cp312/cp313 wheels

Add new Windows SageAttention and xformers wheels

Add flash_attn cp313 wheel for Python 3.13

Add sageattention cp313 wheel for Python 3.13

Add xformers cp313 wheel for Python 3.13

Add xformers cp312 wheel (replaces unreleased py39-none variant)

Update README: add Python 3.13 wheels, replace xformers py39-none with cp312/cp313

Delete xformers-0.0.35+cu130.torch2.11-py39-none-win_amd64.whl

Delete xformers-0.0.35+cu130.torch2.11-py39-none-linux_aarch64.whl

Add Windows cu130 torch2.11 sageattention and xformers wheels

feat: add xformers 0.0.35 for DGX Spark (torch 2.11+cu130, sm_121)

feat: add SageAttention 2.2.0 for DGX Spark (torch 2.11+cu130, py3.12, sm_121a)

feat: add FlashAttention 2.8.4 for DGX Spark (torch 2.11+cu130, py3.12, sm_121)

docs: rewrite README with environment compatibility (DGX Spark / Windows / etc.)

Delete m - Chrome.lnk

Upload sageattention-2.2.0+cu128.torch2.10-cp310-cp310-win_amd64.whl

Upload m - Chrome.lnk

Upload angelslim-0.0.0.dev0+cu128.torch2.10-py3-none-any.whl

Upload deep_gemm-2.3.0+cu128.torch2.10-cp310-cp310-win_amd64.whl

Delete flash_attn-2.8.3-py3-none-any.whl

Upload flash_attn-2.8.3-py3-none-any.whl with huggingface_hub

Upload flash_attn-2.8.3-cp311-cp311-win_amd64.whl

Rename flash_attn-2.8.3+cu130.torch2.9.1.py310.sm121-linux_aarch64-cp310-cp310.whl to flash_attn-2.8.3-cp310-cp310-linux_aarch64.whl

Rename flash_attn-2.8.3+cu130-torch2.9.1-py310-sm121-linux_aarch64-cp310-cp310.whl to flash_attn-2.8.3+cu130.torch2.9.1.py310.sm121-linux_aarch64-cp310-cp310.whl

Rename flash_attn-2.8.3-cu130-sm121-torch2.9.1-py310-cp310-cp310-linux_aarch64.whl to flash_attn-2.8.3+cu130-torch2.9.1-py310-sm121-linux_aarch64-cp310-cp310.whl

Rename flash_attn-2.8.3+130.2.9.1-py310-linux_aarch64-cc121.whl to flash_attn-2.8.3-cu130-sm121-torch2.9.1-py310-cp310-cp310-linux_aarch64.whl

Rename flash_attn-2.8.3+130.2.9.1-py311-linux_aarch64-cc121.whl to flash_attn-2.8.3+130.2.9.1-py310-linux_aarch64-cc121.whl

Rename flash_attn-2.8.3+cu130-torch2.9.1-py311-linux_aarch64-cc121.whl to flash_attn-2.8.3+130.2.9.1-py311-linux_aarch64-cc121.whl

Upload flash_attn-2.8.3+cu130-torch2.9.1-py311-linux_aarch64-cc121.whl

initial commit