Ensure that measure_cold::run_warmup instantiates blocking kernel

Because warm-up runs are executed without use of blocking kernel,
the blocking kernel was not jitted until actual measurements were
collected. The module loading cost incurred during the first run
shows as elevated CPU time noise value for the first measurement
as noted in https://github.com/NVIDIA/nvbench/pull/339

This PR adds `this->block_stream(); this->unblock_stream();` prior
to executing warm-up loop with use of blocking kernel disabled.

This ensures that blocking kernel is instantiated during the warm-up,
but it no other kernel is launched between its launch and stream sync
thus avoiding deadlocking.

nvbench/detail/measure_cold.cuh

@@ -249,6 +249,11 @@ private:
       return;
     }
 
+    // Ensure blocking kernel is loaded during the warmup
+    // Ref: https://github.com/NVIDIA/nvbench/issues/339
+    this->block_stream();
+    this->unblock_stream();
+
     // disable use of blocking kernel for warm-up run
     // see https://github.com/NVIDIA/nvbench/issues/240
     constexpr bool disable_blocking_kernel = true;