Dramatic speedup: pytorch, JSON encoding, uvicorn

[michaelfeil]

• adds new dependency (orjson) for faster response serialization - 300%
• uses torch.inference_mode() and delayed moving to CPU - 10%
• adds uvicorn[standard] - slightly faster 2-5%?

Updates readme

[codecov-commenter]

Welcome to Codecov 🎉

Once merged to your default branch, Codecov will compare your coverage reports and display the results in this comment.

Thanks for integrating Codecov - We've got you covered ☂️