Qulacs

New features & Updates
- Add description about AVX2 in README (241, thanks to y-yu and kwkbtr)
- Add new gate `gate::CP` and `gate::ProbabilisticInstrument` to represent general maps (243 250, thanks to corryvrequan)
- Renew out-dated build guideline in REAMDE (246, thanks to kotamanegi)
- Add fast simulator for noisy circuit (249, thanks to kotamanegi)
- Add concatenate, permutate, and drop qubits in quantum state (245, thanks to corryvrequan)
- Update pybind11 and eigen (253 corryvrequan and 248, thanks to kotamanegi)
- Improve the performance of single-qubit diagonal gate, `Pauli` gate, and `PauliRotation` gate (253, thanks to corryvrequan)
- Move from TravisCI to GitHub Action and resolve wheelbuild issues (256 260 261 263 262 265 266, thanks to y-yu and kwkbtr)
Bug fixes
- Fix a problem that copied parametric circuit does not work (239, thanks to corryvrequan)
- Fix a problem that load function from GPU to CPU does not copy measurement history (242, thanks to corryvrequan)
- Fix double-free issue in parametric gate (244, thanks to corryvrequan and kosukemtr)
- Fix a memory leak in adding a Pauli operator to an observable (251, thanks to corryvrequan)
- Fix a bug in thread-number control (252, thanks to corryvrequan)

This is a hot-fix release that fixes an issue on download URL of eigen (236).
Please also see [v0.1.10 release note](https://github.com/qulacs/qulacs/releases/tag/v0.1.10).

New features & Updates

- Choose random seed for sampling from state (205, thanks to corryvrequan)
- Add argument names and docstrings (209, thanks to corryvrequan)
- Remove `sudo: false` to use Linux infrastructure (206, thanks to y-yu)
- Add support for override of compiler optimization flags (214, thanks to kwkbtr)
- Further SIMD optimization and thread number optimization for Pauli operations (219, thanks to corryvrequan)
- Add Toffoli and Fredkin (220, thanks to corryvrequan)
- Add diagonal gate (221, thanks to corryvrequan)
- Remove experimental qasm load function (223, thanks to corryvrequan)
- Build sdist, bdist_wheel on CI and publish them to PyPI (233, thanks to kwkbtr)
- Add CONTRIBUTING.md (207, thanks to y-yu)

Bug fixes

- Bugfix of bitstring sampling for density matrix (204, thanks to corryvrequan)
- Fix C++ gradient computing coefficients (208, thanks to corryvrequan)
- Fix compiler specification in setup.py to respect environment variables (216 and 229, thanks to kwkbtr)
- Fix Dockerfile (225, thanks to FullteaOfEEIC)
- Fix some commutation properties (232, thanks to akpc)
- Fix typo in example code of PauliOperator (228, thanks to 5c0t-qi)

New features & Updates
* Tutorial update (191, thanks to corryvrequan )
* Implementation on specifying and switching multiple GPUs (194, thanks to puyokw )
* Add a function of multiplying diagonal matrix with python function (197, thanks to puyokw )



Bug fixes
* Check AVX2 support in build script (187, thanks to corryvrequan )
* Apply C++14 flag also for gcc-8 (189, thanks to corryvrequan )
* Add stdio.h include for avoiding Mac OS bugs (190, thanks to corryvrequan )
* Bugfix of projection test (192, thanks to corryvrequan )
* Bugfix for GPU with small targets and many controls (196, thanks to corryvrequan )
* Add `set -o pipefail` to fail CI when failing tests (200, thanks to y-yu )
* Bugfix in trace preserving noise (202, thanks to corryvrequan )

Updates
* SIMD optimization according to Intel AVX2 instructions (181, thanks to corryvrequan)
* Fix definition of “norm” (184, thanks to corryvrequan)

Bugfixes
* Bugfix about GPU index constant (180, thanks to corryvrequan)
* Bugfix of dense matrix gate in gpusim (182, thanks to puyokw)

Bugfixes

* Bugfix for GPU simulator
* Bugfix in get_norm() method, python setup scripts for windows, and matrix gates (thanks to corryvrequan)
* curand flag for windows(149, thanks to shukob and corryvrequan)
* Change ownership of data_cpp to avoid memory leaks in GPU simulator (thanks to corryvrequan)
* Bugfix in build script (thanks to Rogerluo and corryvrequan)