Sema: Only consider key path application for member lookup for SubscriptExprs.

Interactions with the rest of the type checker cause use of functions named `subscript` with backtick-quotes to become ambiguous. This is more fully and robustly solved in master by #9989, but that's too invasive a change for 4.0, and we need a spot fix to prevent source compatibility regressions such as SR-5513.

Author: jckarter

lib/Sema/CSSimplify.cpp

@@ -2891,7 +2891,9 @@ performMemberLookup(ConstraintKind constraintKind, DeclName memberName,
   result.OverallResult = MemberLookupResult::HasResults;
 
   // If we're looking for a subscript, consider key path operations.
-  if (memberName.isSimpleName(getASTContext().Id_subscript)) {
+  if (memberName.isSimpleName(getASTContext().Id_subscript)
+      && (isa<SubscriptExpr>(memberLocator->getAnchor())
+          || isa<KeyPathExpr>(memberLocator->getAnchor()))) {
     result.ViableCandidates.push_back(
         OverloadChoice(baseTy, OverloadChoiceKind::KeyPathApplication));
   }

test/decl/subscript/quoted-subscript-identifier.swift

@@ -0,0 +1,10 @@
+// RUN: %target-swift-frontend -typecheck -verify %s
+// SR-5513
+
+class OTTextView {
+  func `subscript`(_ sender: Any?) {}
+}
+
+class MyTextView: OTTextView {
+  override func `subscript`(_ sender: Any?) { super.`subscript`(sender) }
+}