Fix parallel code gen in RT backend

• Properly use Arc<Mutex<_Option<JoinHandle<_, _>>_>> for task parallel calls.
  • Include Arc clones right before opening async closure (for async calls and for fork-joins).
• Make data parallel fork-joins not task parallel by default (can still be outlined for task parallelism).
• Fix edge case in function inlining.

hercules_cg/src/rt.rs

@@ -594,7 +594,7 @@ impl<'a> RTContext<'a> {
                     (_, true) => {
                         write!(block, "{}", self.clone_arc(id).unwrap())?;
                         format!(
-                            "*async_call_{}.lock().await = Some(::async_std::task::spawn(async move {{ ",
+                            "*async_call_{}.lock().await = ::hercules_rt::__FutureSlotWrapper::new(::async_std::task::spawn(async move {{ ",
                             id.idx(),
                         )
                     }
@@ -1106,7 +1106,7 @@ impl<'a> RTContext<'a> {
             if is_async_call {
                 write!(
                     w,
-                    "let mut async_call_{} = ::std::sync::Arc::new(::async_std::sync::Mutex::new(None));",
+                    "let mut async_call_{} = ::std::sync::Arc::new(::async_std::sync::Mutex::new(::hercules_rt::__FutureSlotWrapper::empty()));",
                     idx,
                 )?;
             } else {
@@ -1382,10 +1382,7 @@ impl<'a> RTContext<'a> {
             && func.schedules[id.idx()].contains(&Schedule::AsyncCall)
         {
             assert!(!lhs);
-            format!(
-                "async_call_{}.lock().await.as_mut().unwrap().await",
-                id.idx(),
-            )
+            format!("async_call_{}.lock().await.inspect().await", id.idx(),)
         } else {
             format!("node_{}", id.idx())
         }