Kernel-launch sync (ADR-0009 D5) and IPCQ drain at inbound (ADR-0023)

- KernelLaunchMsg gains target_start_ns: IO_CPU stamps a global barrier
  (max path latency across every target PE), M_CPU passes it through,
  PE_CPU yields until it before recording pe_exec_start. Every PE in a
  launch begins kernel execution at the same env.now regardless of its
  dispatch path length — eliminates per-PE dispatch-offset artifact in
  cross-PE and cross-cube latency measurements.

- PE_DMA._handle_ipcq_inbound now pays Transaction.drain_ns at the top,
  matching the terminal-drain behavior of ComponentBase._forward_txn for
  every non-IPCQ Transaction. SRC-side tl.send stays fire-and-forget
  (sender doesn't yield on sub_done); tl.recv now blocks until bytes
  have actually drained into its inbox.

- ComponentContext: new compute_path_latency_ns helper + node_overhead_ns
  field populated by GraphEngine.

- tests/test_kernel_launch_sync.py: asserts all PEs in one launch
  produce identical pe_exec_ns for a no-op kernel (zero spread).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/kernbench/components/builtin/pe_cpu.py

@@ -95,6 +95,13 @@ class PeCpuComponent(ComponentBase):
         request = txn.request
         yield from self.run(env, 0)
 
+        # ADR-0009 D5: synchronized launch barrier. If M_CPU stamped a
+        # target_start_ns, wait until then so every PE in this launch
+        # begins pe_exec measurement at the same simulated time.
+        target_start = getattr(request, "target_start_ns", None)
+        if target_start is not None and target_start > env.now:
+            yield env.timeout(float(target_start) - env.now)
+
         kernel_fn = get_kernel(request.kernel_ref.name)
         num_programs = self._derive_num_programs(request)
         kernel_args = self._unpack_kernel_args(request)