Flash Attention: Fast and Memory-Efficient Exact Attention! Repo at https://github.com/Dao-AILab/flash-attention

To try and fix `OSError: CUDA_HOME environment variable is not set. Please set it to your CUDA install root`

Xref conda-forge/conda-forge.github.io#2102

This flash-attn library only runs on Linux with CUDA GPUs if I'm not mistaken.

Needed to compile flash-attn on CUDA 12.0 in conda-forge.

Xref https://conda-forge.org/docs/maintainer/conda_forge_yml/#azure

Only compiling on Compute Capability 8.0 and above, see https://developer.nvidia.com/cuda-gpus. I.e. NVIDIA Ampere generation devices or newer.

This simpler script doesn't have unused features and doesn't set -O3 because our channel defaults are -O2

Co-authored-by: Wei Ji <23487320+weiji14@users.noreply.github.com>

Copied from https://github.com/NVIDIA/cutlass/blob/v3.5.0/LICENSE.txt

Same as 4d8b37c, partially revert 317646a

Silence warnings like:

```
WARNING (flash-attn): dso library package conda-forge/linux-64::libcublas==12.0.1.189=hd3aeb46_3 in requirements/run but it is not used (i.e. it is overdepending or perhaps statically linked? If that is what you want then add it to `build/ignore_run_exports`)
WARNING (flash-attn): dso library package conda-forge/linux-64::libcusparse==12.0.0.76=hd3aeb46_2 in requirements/run but it is not used (i.e. it is overdepending or perhaps statically linked? If that is what you want then add it to `build/ignore_run_exports`)
WARNING (flash-attn): dso library package conda-forge/linux-64::libcusolver==11.4.2.57=hd3aeb46_2 in requirements/run but it is not used (i.e. it is overdepending or perhaps statically linked? If that is what you want then add it to `build/ignore_run_exports`)
```

Trying to reduce CPU load on Azure CI to debug build.