modal: 降级 GPU A100 -> T4 节省成本

deploy: 成功部署 MiniCPM-o-4_5 到 Modal (预编译llama-cpp-python A100)

docs: pre-refactor backup - before tech log integration

fix(deploy): 使用预编译CUDA wheel替代源码编译

重构modal部署使用CUDA编译, streaming修复, cookbook_ref gitignored

[modal] Fix: mmproj single-file, health endpoint, client API consistency

fix(deploy.py): 修复4个问题 - MODEL_DIR shadowing, find_mmproj_files硬编码, MainModel传参, 移除--multimodal flag

fix: CUDA runtime LD_LIBRARY_PATH + CUDA_HOME in Modal deploy; add inference test script

feat(modal): 部署 MiniCPM-o 4.5 到 Modal (llama.cpp + FastAPI)

feat: llama.cpp deploys MiniCPM-o 4.5 on Modal - technical report and client API