Stardist

- Enable model export for [bioimage.io](https://bioimage.io/docs/) model zoo (#171); see new [example notebook](https://github.com/stardist/stardist/blob/b3780c374498c7aad449cb63a86d062a726102a9/examples/other2D/bioimageio.ipynb) how to use
- Add optional `scale` parameter to `model.predict_instances` (181)
- Various new features, improvements, and bug fixes

Not part of the Python package, but important:

- Update documentation (with contributions from psobolewskiPhD and ajinkya-kulkarni)
- Update Dockerfile
- Fix matplotlib configuration issue in example notebooks
- Fix a bioimage.io test issue (FynnBe)
- Update GitHub issue templates
- Fix various GitHub action issues

Maintenance release to support TensorFlow >= 2.6.0, requiring CSBDeep >= 0.6.3 (see [release notes](https://github.com/CSBDeep/CSBDeep/releases/tag/0.6.3)).

Please note the new installation instructions:

- *Unchanged* when using TensorFlow 2.x: `pip install stardist`
- *Updated* when using TensorFlow 1.x: `pip install "stardist[tf1]"`

This is a minor release that fixes a few issues (cf. 57, 154, 156).

Bugfix release (missing files in [`stardist-0.7.0.tar.gz`](https://files.pythonhosted.org/packages/0a/18/e45dd94bfa6d302f42c906571b1c5eb129c7b155bd6514e12e2ed91ee2a0/stardist-0.7.0.tar.gz)).

This only affects (presumably very few) people who can't install from [wheels](https://pypi.org/project/stardist/#files).
Everybody else should already be fine with the `0.7.0` release.

See [StarDist 0.7.0](https://github.com/stardist/stardist/releases/tag/0.7.0) for release notes.

With more than one year in the making, this is one of the biggest releases yet:

* Support for Python 3.9
* New support for multi-class object classification (e.g. for classifying nuclei into different types, details below)
* Lots of additions to increase memory/time efficiency (details below)
* Integration functions for the new [StarDist napari plugin](https://github.com/stardist/stardist-napari)
* Many bug fixes


Object classification

In addition to detecting all *object instances* (e.g. cell nuclei), StarDist can now additionally classify each found object instance into a fixed number of different *object classes* (e.g. cell types). Please see the [example notebook](https://nbviewer.jupyter.org/github/stardist/stardist/blob/master/examples/other2D/multiclass.ipynb) for a demonstration with 2D images, but the feature also works for 3D images.


Efficiency

There are many changes (mostly internal) that increase the overall efficiency. Highlights:

- Non-maximum suppression (NMS) has been refactored and is now considerably faster (now uses kd-trees internally, inspired by PR 40 from GFleishman. Thanks!)
- Better integration of [`edt`](https://github.com/seung-lab/euclidean-distance-transform-3d) to speed up training (we thank william-silversmith for the discussion in #146)
- CNN prediction is now vastly more memory-efficient for large images ([`sparse` flag](https://github.com/stardist/stardist/blob/3451a4f9e7b6dcef91b09635cc8fa78939fb0d29/stardist/models/base.py#L621-L623), enabled by default)

If that's not memory-efficient enough, there's still the option of block-wise processing for very large images ([`predict_instances_big`](https://github.com/stardist/stardist/blob/3451a4f9e7b6dcef91b09635cc8fa78939fb0d29/stardist/models/base.py#L749-L800)).