There is no "Navier-Stokes theorem". There is a famous class of open problems whether Navier-Stokes equations are well-posed (have solutions that don't explode in finite time) for various initial data, but that type of question is completely irrelevant for any practical purposes. I do share the feeling though. As a non-expert, I have no idea what the existing, allegedly practically relevant, formalizations of distributed algorithms actually guarantee.