The full cuStreamSynchronize removal hung the training loop because
cudarc's EventTrackingGuard + check_err need one initial sync to clear
stale events from CUDA Graph capture. After the first call, stream
ordering is sufficient.
- target_ema_update: sync on first call only (target_params_initialized)
- apply_attention_forward: sync on first call only (attention_initialized)
- All subsequent steps: zero sync, pure async kernel dispatch
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>