Dynamiqs

This release fixes some bugs and introduces several improvements. A special thank you to our new contributor vaicecaze, welcome aboard! ⛵

🚀 Features
- GIFs are now 3x faster to create, and the colormap is nicer (see https://github.com/dynamiqs/dynamiqs/pull/723).
<p align="center">
<img width="250" src="https://github.com/user-attachments/assets/f71cdcd3-9dc2-45dc-ba0b-c6fed41eae47">
</p>

- New qubit gates: [`dq.rx()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/rx.html), [`dq.ry()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/ry.html), [`dq.rz()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/rz.html), [`dq.sgate()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/sgate.html), [`dq.tgate()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/tgate.html), [`dq.cnot()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/cnot.html) and [`dq.toffoli()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/toffoli.html).
- New function [`dq.bloch_coordinates()`](https://www.dynamiqs.org/0.2/python_api/utils/quantum_utils/bloch_coordinates.html) which returns the spherical coordinates $(r,\theta,\phi)$ of a state on the Bloch sphere.
- The number operator function [`dq.number()`](https://www.dynamiqs.org/0.2/python_api/utils/operators/number.html) now supports multiple dimensions. For example, for two modes $a$ (dimension 4) and $b$ (dimension 5), the operators $a^\dagger a \otimes I_5$ and $I_4\otimes b^\dagger b$ can be simply obtained with:
python
na, nb = dq.number(4, 5)

- Added an optional `index` argument to [`dq.set_device()`](https://www.dynamiqs.org/0.2/python_api/utils/jax_utils/set_device.html).
- Support for an empty list of jump operators in [`dq.lindbladian()`](https://www.dynamiqs.org/0.2/python_api/utils/quantum_utils/lindbladian.html) and its vectorized version [`dq.slindbladian()`](https://www.dynamiqs.org/0.2/python_api/utils/vectorization/slindbladian.html), allowing the construction of a Lindbladian from a Hamiltonian only.

🐛 Bugs
- Fixed a bug that occurred when mixing subplots with colorbars. Everything should look good now (see https://github.com/dynamiqs/dynamiqs/pull/761).

📖 Documentation
- The documentation now includes a version selector next to the search bar, to switch between multiple versions of the library (`dev` = latest commit on the `main` branch of the repository, `dev-qarray` = what we're cooking for the next big release):
<p align="center">
<img width="250" src="https://github.com/user-attachments/assets/7deab806-66f8-4315-8425-cd99a1ed1e33">
</p>

📦 Other changes
- We have a new smoother logo from our logo-master gautierronan. Can you spot the seven differences from the previous one (on the left)?
<p align="center">
<img width="250" src="https://github.com/user-attachments/assets/2cd7f583-7c8b-433c-a0c4-c38659e9e770">
</p>

**Full Changelog**: https://github.com/dynamiqs/dynamiqs/compare/v0.2.1...v0.2.2

This release primarily addresses installation issues with the `uv` package manager. The library's official name is now `Dynamiqs` instead of `dynamiqs`, although the repository and Python package will still be named `dynamiqs`.

🚀 Features
- Add the `save_extra` option for the `dq.sepropagator()` and `dq.mepropagator()` solvers.
- Support double precision on GPU for `dq.wigner()`.

📖 Documentation
- Add analytical gradient calculations results to the README and documentation basic example.

📦 Other changes
- Change the default plot font of `dq.plot.mplstyle()` from `serif` to `sans-serif`.

**Full Changelog**: https://github.com/dynamiqs/dynamiqs/compare/v0.2.0...v0.2.1

First release based on JAX. 🪄 Congratulations and special thanks to our new contributors: lucasoeth, RemiRousseau, bsaoptima, gaspardbb, etienney and dkweiss31!

Latest version of the library based on PyTorch. From the next release, the library will be powered by JAX. 🚀