Alternative FREEC usage from a container

[kirilllzaitsev]

The goal is to make FREEC experiments simpler without setting up C++ make and miscellaneous libraries on the end-user's operating system.
Docker allows developers to bake a container, ready for execution by users with a one-liner command.

Pros:

• users don't need to mirror the developer's setup (make, g++, their versions, etc.)
• runs on any OS

Cons:

• users have to install Docker (this is substantially easier than that listed above)
• need to keep track of folder locations: as per the README, users make a minor modification to a configuration file and have to use the instructions in the README carefully to get the results

Image on the Dockerhub should be rehosted to a BoevaLab repository.

Q&A

Versioning
Dockerhub provides the same means of image versioning as Github for code. It could be automated, but for now, it is the developer's responsibility to keep the versions of code and an image built from this code the same.

Running with data split across multiple folders (in the local filesystem)
--mount option in the docker run command can be used multiple times, one for each folder where the required data resides