Call jit_post_compile_hook within Inductor Triton Kernel compile path (#83)

Summary:
Pull Request resolved: #83

X-link: pytorch/pytorch#161443

Since Inductor skips JIT compilation for Triton kernels, we need to manually invoke `knobs.runtime.jit_post_compile_hook` if one exists. Here, we do this to enable Tritonparse to extract launch metadata from Inductor launched kernels. We can control whether or not Inductor will run the hook with a new `TORCHINDUCTOR_RUN_JIT_POST_COMPILE_HOOK=1 ` config variable.

Reviewed By: davidberard98, FindHao

Differential Revision: D80624932

fbshipit-source-id: 57cd14f3facf053fe1c02b15f04f0ea4c1b68198

Author: NikhilAPatel

tritonparse/structured_logging.py

@@ -875,12 +875,14 @@ def __call__(
                 return True
 
         # Get the current launch_metadata function if it exists
-        current_launch_metadata = getattr(fn.jit_function, "launch_metadata", None)
+        function = getattr(fn, "jit_function", fn)
+
+        current_launch_metadata = getattr(function, "launch_metadata", None)
         if current_launch_metadata is not None:
             log.warning(
                 f"fn {fn} launch_metadata is not None: {current_launch_metadata}. It will be overridden by tritonparse."
             )
-        fn.jit_function.launch_metadata = add_launch_metadata
+        function.launch_metadata = add_launch_metadata
         return True