Bump flash-attn from 2.4.2+7e2dd4d to 2.8.3.post1

Bumps [flash-attn](https://github.com/Dao-AILab/flash-attention) from 2.4.2+7e2dd4d to 2.8.3.post1.
- [Release notes](https://github.com/Dao-AILab/flash-attention/releases)
- [Commits](https://github.com/Dao-AILab/flash-attention/commits/v2.8.3.post1)

---
updated-dependencies:
- dependency-name: flash-attn
  dependency-version: 2.8.3.post1
  dependency-type: direct:development
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

Author: dependabot[bot]

pyproject.toml

@@ -97,7 +97,7 @@ enflame = [
     "torchaudio==2.10.0+cpu",
     "torchvision==0.25.0+cpu",
     "torch-gcu==2.10.0+3.7.20260408",
-    "flash-attn==2.7.2+torch.2.9.1.gcu.3.4.20260323",
+    "flash-attn==2.8.3.post1",
 ]
 
 hygon = [
@@ -115,7 +115,7 @@ kunlunxin = [
     "apex==0.1",
     "benchflow==1.0.0",
     "colorama==0.4.6",
-    "flash_attn==2.4.2+7e2dd4d",
+    "flash_attn==2.8.3.post1",
     "hydrax==0.1.0",
     "hyperparameter==0.5.6",
     "psutil==6.1.0",
@@ -133,7 +133,7 @@ metax = [
     "torch==2.8.0+metax3.7.2.0",
     "torchaudio==2.4.1+metax3.7.2.0",
     "torchvision==0.15.1+metax3.7.2.0",
-    "flash_attn==2.6.3+metax3.7.2.0torch2.8",
+    "flash_attn==2.8.3.post1",
 ]
 
 mthreads-musa436 = [