[llvm] [orc-rt] Implement Windows target triple detection (PR #224514)

[Lang Hames via llvm-commits]

================
@@ -93,18 +93,16 @@ class OptionParser {
                          OptionKind Kind = OptionKind::Value,
                          std::optional<char> ShortName = std::nullopt) {
     Val = DefaultVal;
-    Opts.push_back({.Name = std::string(Name),
-                    .ShortName = std::move(ShortName),
-                    .Desc = std::string(Desc),
-                    .Kind = Kind,
-                    .Default = [&Val, DV = DefaultVal]() { Val = DV; },
-                    .FromString = [&Val, OptName = std::string(Name)](
-                                      std::string_view S) -> Error {
+    Opts.push_back({std::string(Name), std::move(ShortName), std::string(Desc),
+                    Kind, [&Val, DV = DefaultVal]() { Val = DV; },
+                    [&Val, OptName = std::string(Name)](
+                        std::string_view S) -> orc_rt::Error {
                       if (auto V = detail::parseValue<T>(S)) {
                         Val = *V;
-                        return Error::success();
+                        return orc_rt::Error::success();
----------------
lhames wrote:

`orc_rt::`-qualification shouldn't be needed here, right?

https://github.com/llvm/llvm-project/pull/224514