[Fix] Preserve DeepEP layout events in synchronous dispatch

[jayhenry]

Stack #2179 (bottom to top):

1. [Fix] Preserve DeepEP layout events in synchronous dispatch #2177 fix/deepep-layout-event → main ← you are here
2. [Feature] Add GLM-5.3-Flash F0: 25B cropped reference checkpoint builder #2105 feat/glm53flash-materialize-full-f0 → fix/deepep-layout-event
3. [Feature] Add GLM-5.3-Flash F3: Kimi Delta Attention (KDA) #2106 feat/glm53flash-f3-kda → feat/glm53flash-materialize-full-f0
4. [Feature] Add GLM-5.3-Flash F4: mHC four-stream residual #2107 feat/glm53flash-f4-mhc → feat/glm53flash-f3-kda
5. [Feature] Add GLM-5.3-Flash F5: NoPE DSA + KPool indexer + clamped SwiGLU #2108 feat/glm53flash-f5-nope-dsa → feat/glm53flash-f4-mhc
6. [Feature] Add GLM-5.3-Flash F1: VL data preprocessing pipeline #2109 feat/glm53flash-f1-vl-data → feat/glm53flash-f5-nope-dsa
7. [Feature] Add GLM-5.3-Flash F2: vision tower + projector (eager) #2110 feat/glm53flash-f2-vision-tower → feat/glm53flash-f1-vl-data
8. [Fix] Correct generic MTP future embeddings and empty-shard loss #2178 fix/mtp-local-future-embeddings → feat/glm53flash-f2-vision-tower
9. [Feature] Add GLM-5.3-Flash F6 core: text model + MTP + compose model #2111 feat/glm53flash-f6-text-moe → fix/mtp-local-future-embeddings

Base is main. Review only this PR's own diff.

---

Summary

• Keep the DeepEP layout event when dispatch is synchronous, so the small layout tensors can pass their communication stream handoff to dispatch without a blocking compute stream transition.
• Check that payload sized dispatch and combine buffers remain immediately reusable after the call.

This is the first layer of the GLM-5.3-Flash stack and is based on main. Its implementation and test are independent of the model layers above it.

Tests

• tests/module/dispatcher/test_deepep.py::TestMoETorchAll2AllDispatcher::test_sync_dispatch_and_combine_buffers_are_reusable_once_freed passed with two GPUs as part of the 17 passing targeted GPU tests.