[llvm] [KernelInfo] Implement new LLVM IR pass for GPU code analysis (PR #102944)

[Matt Arsenault via llvm-commits]

================
@@ -1354,6 +1354,12 @@ bool TargetTransformInfo::hasActiveVectorLength(unsigned Opcode, Type *DataType,
   return TTIImpl->hasActiveVectorLength(Opcode, DataType, Alignment);
 }
 
+void TargetTransformInfo::collectLaunchBounds(
----------------
arsenm wrote:

Maybe collectKernelLaunchBounds? 

https://github.com/llvm/llvm-project/pull/102944