Add ChatGLM Model Support

[Qubitium]

Add ChatGLM Model loading. Code adapted from vllm main.

Notable Changes:

1. model_config.get_num_kv_heads() and .get_total_num_kv_heads() ported from vllm to correctly caculate kv heads from model config json. Found that chatglm with sglang main code retrieved wrong kv heads from config causing kv cache to contain the wrong shape.
2. New EntryClassRemapping property added to model entry definition to help with future compat. Chatglm has ChatGLMModel set in config.json when model loader needs ChatGLMForCausalLM.

Remapping code:

	    # compat: some models such as chatglm has incorrect class set in config.json
            # usage: [ tuple("From_Entry_Class_Name": EntryClass), ]
            if hasattr(module, "EntryClassRemapping") and isinstance(module.EntryClassRemapping, list):
                for remap in module.EntryClassRemapping:
                    if isinstance(remap, tuple) and len(remap) == 2:
                        model_arch_name_to_cls[remap[0]] = remap[1]

Usage:

EntryClass = ChatGLMForCausalLM
# compat: glm model.config class == ChatGLMModel
EntryClassRemapping = [("ChatGLMModel", ChatGLMForCausalLM)]

TESTS:

• PASSED ChatGLM
• PASSED Non-ChatGLM: regression test for new get_num_kv_heads
• PASSED TP=2
• PASSED DP=2