Tests Occasionally Fail Due to Gateway Timeout #80
Description
The current test cases rely on downloading data from external sources. The ground-truth datasets are downloaded from zenodo during test setup to compare the output of the Bonsai workflow to expected data that are pre-generated. While this reduces the number of python dependencies and memory space required to generate the expected data during each run, the requests to download the dataset from zenodo can fail, leading to the GitHub action workflow to fail.
One possible solution to this would be to design a workflow that can make use of the GitHub action cache, where in principle we could generate the expected outputs/dependencies for tests, examples, etc. only once and then carry the cache forward throughout the workflow.