Create docker container for XBT environment

Another way to create a reproducible research environment would be to create a docker container with the relevant environment in it. We could create a dockerfile then demonstrate how to run it. This could be tested running on compute environments such as:
* AWS batch
* Azure Batch
* EC2 instance
* Azure compute

Start by identifying the dependencies and runtime environment required by XBTs_classification, then define the Dockerfile and its run instructions. Done means the container is reproducible, its usage is documented, and it has been tested on the listed AWS Batch, Azure Batch, EC2, and Azure compute environments.

Written by the indexing model from the issue text.