flash_attn-2.6.3 installation crash

**Trouble**:

During installation with `pip install -e .`, I encountered the following error:

```
Obtaining file:///omniparse
Installing build dependencies ... done
Checking if build backend supports build_editable ... done
Getting requirements to build editable ... done
Preparing editable metadata (pyproject.toml) ... done
Collecting Pillow<11.0.0,>=10.1.0 (from omniparse==0.0.1)
Downloading pillow-10.4.0-cp310-cp310-manylinux_2_28_x86_64.whl.metadata (9.2 kB)
Collecting art<7.0,>=6.2 (from omniparse==0.0.1)
Downloading art-6.2-py3-none-any.whl.metadata (69 kB)
Collecting beautifulsoup4<5.0.0,>=4.12.3 (from omniparse==0.0.1)
Using cached beautifulsoup4-4.12.3-py3-none-any.whl.metadata (3.8 kB)
Collecting fastapi<0.112.0,>=0.111.0 (from omniparse==0.0.1)
Using cached fastapi-0.111.1-py3-none-any.whl.metadata (26 kB)
Collecting filetype<2.0.0,>=1.2.0 (from omniparse==0.0.1)
Downloading filetype-1.2.0-py2.py3-none-any.whl.metadata (6.5 kB)
Collecting flash-attn<3.0.0,>=2.5.9 (from omniparse==0.0.1)
Downloading flash_attn-2.6.3.tar.gz (2.6 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.6/2.6 MB 61.2 MB/s eta 0:00:00
Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> [6 lines of output]
Traceback (most recent call last):
File "", line 2, in
File "", line 34, in
File "/tmp/pip-install-hg68yz3k/flash-attn_eb201cf3b3c94e82bea7ad772cbcb1e2/setup.py", line 21, in
import torch
ModuleNotFoundError: No module named 'torch'
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.
```

**Solution**:

Installed torch manually for my cuda-version (11.8) with conda:
`conda install pytorch==2.2.2 torchvision==0.17.2 torchaudio==2.2.2 pytorch-cuda=11.8 -c pytorch -c nvidia`

Note that torch==2.2.2 is explicitly required in the pyproject.toml for compatibility reasons.

Cheers!