Merge pull request #33 from SharpAI/feat/deepseek-v4

feat: add DeepSeek-V4 model support

Author: solderzzc

Libraries/MLXLLM/LLMModelFactory.swift

@@ -46,6 +46,7 @@ public enum LLMTypeRegistry {
         "openelm": create(OpenElmConfiguration.self, OpenELMModel.init),
         "internlm2": create(InternLM2Configuration.self, InternLM2Model.init),
         "deepseek_v3": create(DeepseekV3Configuration.self, DeepseekV3Model.init),
+        "deepseek_v4": create(DeepseekV4Configuration.self, DeepseekV4Model.init),
         "granite": create(GraniteConfiguration.self, GraniteModel.init),
         "granitemoehybrid": create(
             GraniteMoeHybridConfiguration.self, GraniteMoeHybridModel.init),