[BUILD] Make sure always build_ext first (#1819)

The third-party backend might install its python package to the
`triton/third_party` python package during the build process. But the
`build_py` could be executed before the `build_ext`, and then `build_py`
would only copy the `packages` defined in the `setup.py` w/o the
third-party related packages as the third-party backend has not been
built, which is triggered by `build_ext`. Therefore, this PR refined the
build order a little bit to ensure `build_ext` always happens before
`build_py`.

python/setup.py

@@ -13,6 +13,7 @@ from typing import NamedTuple
 
 from setuptools import Extension, setup
 from setuptools.command.build_ext import build_ext
+from setuptools.command.build_py import build_py
 
 
 # Taken from https://github.com/pytorch/pytorch/blob/master/tools/setup_helpers/env.py
@@ -146,6 +147,11 @@ def download_and_copy_ptxas():
 
 # ---- cmake extension ----
 
+class CMakeBuildPy(build_py):
+    def run(self) -> None:
+        self.run_command('build_ext')
+        return super().run()
+
 
 class CMakeExtension(Extension):
     def __init__(self, name, path, sourcedir=""):
@@ -280,7 +286,7 @@ setup(
     ],
     include_package_data=True,
     ext_modules=[CMakeExtension("triton", "triton/_C/")],
-    cmdclass={"build_ext": CMakeBuild},
+    cmdclass={"build_ext": CMakeBuild, "build_py": CMakeBuildPy},
     zip_safe=False,
     # for PyPI
     keywords=["Compiler", "Deep Learning"],