Add AWQ mapping for GlmMoeDsaForCausalLM (GLM-5) (vllm-project#2418)

Closes vllm-project#2412 (part of vllm-project#1442)

GLM-5 (`GlmMoeDsaForCausalLM`) uses Multi-head Latent Attention
identical to DeepSeek V3 — same projection layer names (`q_a_proj`,
`kv_a_proj_with_mqa`, etc.) and same MoE structure. Reuses
`_deepseek_mappings`.

Also moves `Glm4MoeForCausalLM` to its correct alphabetical position in
the registry.

SUMMARY:
Add GLM-5 (GlmMoeDsaForCausalLM) to AWQ_MAPPING_REGISTRY using
_deepseek_mappings. GLM-5's MLA layer names are identical to DeepSeek
V3. Also fixes alphabetical ordering of Glm4MoeForCausalLM.

TEST PLAN:
Registry-only change (no logic modified). Verified that GLM-5 layer
names (q_a_proj, kv_a_proj_with_mqa, kv_a_layernorm, kv_b_proj, o_proj)
match the patterns in _deepseek_mappings by inspecting the
GlmMoeDsaForCausalLM source in transformers.

Signed-off-by: gillesturpin <turpingilles@orange.fr>
Co-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>
Co-authored-by: HDCharles <39544797+HDCharles@users.noreply.github.com>
Signed-off-by: yiliu30 <yi4.liu@intel.com>

Author: 3 people

src/llmcompressor/modifiers/awq/mappings.py

@@ -251,6 +251,8 @@ class AWQMapping:
     "Gemma2ForCausalLM": _gemma_mappings,
     "Gemma3ForCausalLM": _gemma_mappings,
     "Gemma3ForConditionalGeneration": _gemma_mappings,
+    "Glm4MoeForCausalLM": _default_mappings,
+    "GlmMoeDsaForCausalLM": _deepseek_mappings,
     "LlamaForCausalLM": _default_mappings,
     "Llama4ForConditionalGeneration": _default_mappings,
     "Mistral3ForConditionalGeneration": _default_mappings,
@@ -264,7 +266,6 @@ class AWQMapping:
     "Qwen3ForCausalLM": _default_mappings,
     "Qwen3MoeForCausalLM": _moe_default_mappings,
     "Qwen3NextForCausalLM": _qwen3_next_moe_mappings,
-    "Glm4MoeForCausalLM": _default_mappings,
     "SeedOssForCausalLM": _default_mappings,
     "Ernie4_5_MoeForCausalLM": _default_mappings,
 }