fix(llm): local Ollama provider falls back to no-think when model lacks thinking support

feat(llm): local Ollama provider (ollama_local, no auth) + scenario-neutral agent directive + gpt-oss model toggle

feat(cp4): port cloud providers (openai/anthropic/gemini/ollama) into proteus

feat(cp4): provider factory (make_provider) with lazy registry + fake path

fix(cp2): FakeProvider records dict copies + test empty-responses guard

feat(cp2): providers base + thinking_utils + FakeProvider