JAX-CFD in the Mosaic benchmark suite

We just released **Mosaic** ([github.com/pasteurlabs/mosaic](https://github.com/pasteurlabs/mosaic)), a benchmark for *differentiable* physics solvers. Each solver runs in a container behind a uniform `apply` / `vjp` interface, so one harness compares gradient accuracy and cost across JAX, PyTorch, Julia, and C++-adjoint codes on shared tasks.

**JAX-CFD is one of the benchmarked solvers** on the ns-grid domain ([results](https://docs.pasteurlabs.ai/projects/mosaic/stable/docs/results.html) if you're interested).

We did our best to do it justice, but we don't know your code as well as you do and have surely made suboptimal calls. For example, we use the semi-implicit finite-difference (MAC-grid) stepper with the default advection scheme and spectral pressure projection, and differentiate through it with JAX autodiff. If you'd pick a different advection method, CFL or step size, or projection tolerance for accurate gradients, we'd love to know.

The setup is self-contained under [`mosaic/tesseracts/navier-stokes-grid/jax-cfd/`](https://github.com/pasteurlabs/mosaic/tree/main/mosaic/tesseracts/navier-stokes-grid/jax-cfd/). If something looks odd, corrections or a PR are very welcome, here on GitHub or on the [Tesseract Forum](https://si-tesseract.discourse.group/).

Mostly we wanted you to know your work is represented, and to get it right.

Cheers,
Dion on behalf of the Mosaic authors

Review the external setup under mosaic/tesseracts/navier-stokes-grid/jax-cfd/ and compare its semi-implicit MAC-grid stepper, advection scheme, CFL or step size, and spectral pressure projection with JAX-CFD's documented usage. Done means identifying a concrete correction or improvement and proposing it as a focused change or PR; the issue names no repository file or test to modify.

Written by the indexing model from the issue text.