Fix vote and shuffle shader instructions on AMD GPUs (#5540)

* Move shuffle handling out of the backend to a transform pass

* Handle subgroup sizes higher than 32

* Stop using the subgroup size control extension

* Make GenerateShuffleFunction static

* Shader cache version bump

src/Ryujinx.Graphics.Shader/CodeGen/Glsl/Instructions/InstGenShuffle.cs

@@ -0,0 +1,25 @@
+using Ryujinx.Graphics.Shader.StructuredIr;
+using Ryujinx.Graphics.Shader.Translation;
+
+using static Ryujinx.Graphics.Shader.CodeGen.Glsl.Instructions.InstGenHelper;
+
+namespace Ryujinx.Graphics.Shader.CodeGen.Glsl.Instructions
+{
+    static class InstGenShuffle
+    {
+        public static string Shuffle(CodeGenContext context, AstOperation operation)
+        {
+            string value = GetSoureExpr(context, operation.GetSource(0), AggregateType.FP32);
+            string index = GetSoureExpr(context, operation.GetSource(1), AggregateType.U32);
+
+            if (context.HostCapabilities.SupportsShaderBallot)
+            {
+                return $"readInvocationARB({value}, {index})";
+            }
+            else
+            {
+                return $"subgroupShuffle({value}, {index})";
+            }
+        }
+    }
+}