Fix #4135: Keep bool-to-int conditional in its local

Once comp(x != 0) is removed for a bool x, `int n = b ? 1 : 0` becomes a
plain copy of the bool, and copy propagation replaced every load of n with
`(b ? 1 : 0)`. Refuse copies between bool and non-bool variables rather
than requiring identical types, which would also block legitimate copies
such as dynamic -> object or Derived[] -> Base[].

ICSharpCode.Decompiler.Tests/TestCases/Pretty/TypeAnalysisTests.cs

@@ -377,6 +377,23 @@ namespace ICSharpCode.Decompiler.Tests.TestCases.Pretty
 		{
 			return (int)a == 0;
 		}
+
+		public static byte BoolToIntLocalUsedTwice(byte b, bool flag)
+		{
+			int num = (flag ? 1 : 0);
+			return (byte)(b + num >> num);
+		}
+
+		public static int BoolToIntLocalAsLoopBound(object[] array, bool skipFirst)
+		{
+			int num = (skipFirst ? 1 : 0);
+			int num2 = array.Length - 1;
+			while (num2 >= num && array[num2] == null)
+			{
+				num2--;
+			}
+			return num2;
+		}
 	}
 
 	public enum UIntEnum : uint

ICSharpCode.Decompiler/IL/Transforms/CopyPropagation.cs

@@ -124,6 +124,10 @@ namespace ICSharpCode.Decompiler.IL.Transforms
 					{
 						return false; // non-local code move might change semantics when there's split variables
 					}
+					if (v.Type.IsKnownType(KnownTypeCode.Boolean) != target.Type.IsKnownType(KnownTypeCode.Boolean))
+					{
+						return false;
+					}
 					switch (v.Kind)
 					{
 						case VariableKind.Parameter: