Add Novita LLM and Embedding Provider

[Alex-wuhu]

Summary

• Adds novita provider support for both LLM and Embedding.
• Aligns Novita provider behavior with existing OpenAI-compatible providers.
• Uses OpenAI-compatible Novita endpoint: https://api.novita.ai/openai.
• Supports NOVITA_API_KEY authentication.

Key Fixes (follow-up)

• LLM default model updated to deepseek/deepseek-v3.2 (Novita rule).
• Added response_format pass-through in NovitaLLM.generate_response.
• Added response_callback support in Novita LLM config/runtime (parity with OpenAI provider).
• LLM env var handling aligned with embedding side:
  • prefer NOVITA_API_URL
  • keep NOVITA_API_BASE for compatibility with deprecation warning.
• Unified http_client_proxies typing in Novita configs.
• Embedding default model updated to qwen/qwen3-embedding-0.6b with default dimensions 1024.

Files Changed

• mem0/llms/novita.py
• mem0/embeddings/novita.py
• mem0/configs/llms/novita.py
• mem0/configs/embeddings/novita.py
• tests/llms/test_novita.py
• tests/embeddings/test_novita_embeddings.py

Verification

• uv run --with pytest python -m pytest -q tests/llms/test_novita.py tests/embeddings/test_novita_embeddings.py
• Result: all tests passed.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[MontaEllis8]

✅ Code Review 通过

代码审查通过，建议合并。