[llvm] [DirectX] only allow intrinsics defined in DXIL.td (PR #128613)

[Farzon Lotfi via llvm-commits]

================
@@ -286,7 +286,7 @@ declare <3 x half> @llvm.dx.nclamp.v3f16(<3 x half>, <3 x half>, <3 x half>)
 declare <4 x float> @llvm.dx.nclamp.v4f32(<4 x float>, <4 x float>, <4 x float>)
 declare <2 x double> @llvm.dx.nclamp.v2f64(<2 x double>, <2 x double>, <2 x double>)
 declare <4 x i32> @llvm.dx.sclamp.v4i32(<4 x i32>, <4 x i32>, <4 x i32>)
-declare <3 x i16> @llvm.dx.uclamp.v3i32(<3 x i16>, <3 x i32>, <3 x i16>)
+declare <3 x i16> @llvm.dx.uclamp.v3i16(<3 x i16>, <3 x i16>, <3 x i16>)
----------------
farzonl wrote:

Well one its a typo. but it broke when I changed to error by default.There was no uclamp.v3i32 usage only a declare. There was a uclamp.v3i16 usage but no declare.

If we had a use then after the `DXILIntrinsicExpansion` pass ran it would have cleanup the decalres aswell. As is though this intrinsic  declaration lives on and triggered our error because there is n uclamp lowering to a dxil op.

https://github.com/llvm/llvm-project/pull/128613