improve parallel benchmark speed

See tokio-rs/tokio#2401

Signed-off-by: 20k-ultra <[email protected]>

profiler/src/lib.rs

@@ -23,8 +23,11 @@ pub async fn benchmark(domain: &str, concurrency: usize) -> Result<Vec<Benchmark
         let domain_copy = domain.clone();
         tasks.push(handle.spawn(async move { run_benchmark(domain_copy).await }));
     }
-    let results = futures::future::join_all(tasks).await;
-    results.into_iter().map(Result::unwrap).collect()
+    let mut results = Vec::new();
+    for handle in tasks {
+        results.push(handle.await?.unwrap());
+    }
+    Ok(results)
 }
 
 async fn run_benchmark(domain: Arc<String>) -> Result<Benchmark, io::Error> {