Add test for template string escaping fix and force escaping for control characters

Co-authored-by: jakebailey <5341706+jakebailey@users.noreply.github.com>

Author: Copilot

internal/printer/printer_test.go

@@ -29,6 +29,7 @@ func TestEmit(t *testing.T) {
 		{title: "BooleanLiteral#2", input: `false`, output: `false;`},
 		{title: "NoSubstitutionTemplateLiteral", input: "``", output: "``;"},
 		{title: "NoSubstitutionTemplateLiteral#2", input: "`\n`", output: "`\n`;"},
+		{title: "NoSubstitutionTemplateLiteral#3", input: "`\u001f`", output: "`\\u001F`;"},
 		{title: "RegularExpressionLiteral#1", input: `/a/`, output: `/a/;`},
 		{title: "RegularExpressionLiteral#2", input: `/a/g`, output: `/a/g;`},
 		{title: "NullLiteral", input: `null`, output: `null;`},

internal/printer/utilities.go

@@ -205,6 +205,21 @@ func canUseOriginalText(node *ast.LiteralLikeNode, flags getLiteralTextFlags) bo
 		}
 	}
 
+	// For template literals, check if they contain characters that need escaping
+	if node.Kind == ast.KindNoSubstitutionTemplateLiteral || 
+	   node.Kind == ast.KindTemplateHead ||
+	   node.Kind == ast.KindTemplateMiddle ||
+	   node.Kind == ast.KindTemplateTail {
+		text := node.TemplateLiteralLikeData().Text
+		for _, ch := range text {
+			// Check if this character needs escaping according to the TypeScript PR #60303 fix
+			// Characters in range \u0000-\u001f (excluding \u000a which is handled separately) should be escaped
+			if ch <= '\u001f' && ch != '\n' {
+				return false // Force escaping path
+			}
+		}
+	}
+
 	// Finally, we do not use the original text of a BigInt literal
 	// TODO(rbuckton): The reason as to why we do not use the original text for bigints is not mentioned in the
 	// original compiler source. It could be that this is no longer necessary, in which case bigint literals should