[Feature] Add GLM-5.3-Flash F4: mHC four-stream residual

[jayhenry]

Stack (bottom to top):

1. [Feature] Add GLM-5.3-Flash F0: 25B cropped reference checkpoint builder #2105 feat/glm53flash-materialize-full-f0 → main
2. [Feature] Add GLM-5.3-Flash F3: Kimi Delta Attention (KDA) #2106 feat/glm53flash-f3-kda → feat/glm53flash-materialize-full-f0
3. [Feature] Add GLM-5.3-Flash F4: mHC four-stream residual #2107 feat/glm53flash-f4-mhc → feat/glm53flash-f3-kda ← you are here
4. [Feature] Add GLM-5.3-Flash F5: NoPE DSA + KPool indexer + clamped SwiGLU #2108 feat/glm53flash-f5-nope-dsa → feat/glm53flash-f4-mhc
5. [Feature] Add GLM-5.3-Flash F1: VL data preprocessing pipeline #2109 feat/glm53flash-f1-vl-data → feat/glm53flash-f5-nope-dsa
6. [Feature] Add GLM-5.3-Flash F2: vision tower + projector (eager) #2110 feat/glm53flash-f2-vision-tower → feat/glm53flash-f1-vl-data
7. [Feature] Add GLM-5.3-Flash F6 core: text model + MTP + compose model #2111 feat/glm53flash-f6-text-moe → feat/glm53flash-f2-vision-tower

Base is #2106's branch (layer 2). Review only this PR's own diff.

---

Summary

Stack layer 3/7 of GLM-5.3-Flash support (base: layer 2, F3 KDA).

Adds xtuner/v1/module/decoder_layer/mhc.py (MHCConfig / hc_split_sinkhorn / hc_pre / hc_post / unshard_hc_params) and xtuner/v1/ops/hc_post.py (fused Triton hc_post_fused), per doc/xtuner_glm5p3flash_design.md F4. The core math is ported from xtuner's dsv4 branch (DeepSeek-V4's Hyper-Connections) into the model-agnostic public location the design doc calls for, dropping V4's XTUNER_V4_HF_PARITY global toggle and the unported 721-line TileKernels backend (left as a documented NotImplementedError gap rather than a blind, unvalidated port).

Adds xtuner/v1/model/moe/glm53/decoder_layer.py: Glm53DenseDecoderLayer/Glm53MoEDecoderLayer, which only override the pre/post-forward seams so the existing EP/dispatcher/domino-micro-batch pipeline stays untouched — the mHC residual rides through as an opaque _MHCResidual payload in place of a plain Tensor. mhc_cfg=None degrades both layers to ordinary residual math unchanged, for reuse by the MTP layer later in the stack.

Also fixes a real bug surfaced while testing: KDA's dt_bias/A_log were left torch.empty-uninitialized; now initialized to match HF's Glm5NextTextForgetGate._init_weights policy.

Test Plan

26 tests across tests/model/test_glm53_mhc.py (Sinkhorn math + exact HF parity in fp32, bf16 parity within ULP tolerance), tests/ops/test_hc_post.py (fused kernel vs eager reference, forward/backward/compile), and tests/model/test_glm53_decoder_layer.py (both decoder layers, mhc_cfg=None passthrough equivalence, gradient flow).