ci: add self-hosted Windows build workflow

- Add _build_windows_self_host.yml for self-hosted Windows wheel builds.
- Integrate self-hosted Windows build job into main build.yml workflow.
- Update create_matrix.py to include and enable Windows self-hosted build matrix.
- Implement comprehensive cleanup steps in the self-hosted runner workflow to ensure a clean state for subsequent runs.

create_matrix.py

@@ -195,6 +195,34 @@ WINDOWS_CODEBUILD_MATRIX = {
     ],
 }
 
+WINDOWS_SELF_HOSTED_MATRIX = {
+    "flash-attn-version": [
+        # "2.6.3",
+        # "2.7.4",
+        "2.8.3",
+    ],
+    "python-version": [
+        # "3.10",
+        # "3.11",
+        "3.12",
+        # "3.13",
+    ],
+    "torch-version": [
+        # "2.5.1",
+        # "2.6.0",
+        # "2.7.1",
+        # "2.8.0",
+        "2.9.1",
+    ],
+    "cuda-version": [
+        # "12.4",
+        # "12.6",
+        "12.8",
+        # "12.9",
+        # "13.0",
+    ],
+}
+
 
 def main():
     print(
@@ -206,8 +234,8 @@ def main():
                 "linux_arm64": False,
                 # "linux_arm64": LINUX_ARM64_MATRIX,
                 #
-                # "linux_self_hosted": False,
-                "linux_self_hosted": LINUX_SELF_HOSTED_MATRIX,
+                "linux_self_hosted": False,
+                # "linux_self_hosted": LINUX_SELF_HOSTED_MATRIX,
                 #
                 "linux_arm64_self_hosted": False,
                 # "linux_arm64_self_hosted": LINUX_ARM64_SELF_HOSTED_MATRIX,
@@ -215,6 +243,9 @@ def main():
                 "windows": False,
                 # "windows": WINDOWS_MATRIX,
                 #
+                # "windows_self_hosted": False,
+                "windows_self_hosted": WINDOWS_SELF_HOSTED_MATRIX,
+                #
                 "windows_code_build": False,
                 # "windows_code_build": WINDOWS_CODEBUILD_MATRIX,
                 #