doc: explain `.datalad/make/priority` file

[mslw]

Help for make contains:

  --label LABEL         Label of the computation. This is a user defined name that is used to identify and prioritize
                        computations, if more than one computation is registered for a file. If no label is provided, the
                        template name will be used. (Prioritization is done by reading the git configuration
                        `datalad.make.priority` (which should contain a comma-separated list of labels). If this
                        configuration key does not exist, the priority list is read from the file
                        `$DATASET/.datalad/make/priority`. If that does not exist either, a random computation is chosen.).

We have a "files" section in the docs. The .datalad/make/priority file should be included, too.

[mslw]

While at it, could be good to verify that it is indeed random. E.g. there is this in the code and I am not sure what that implies.

datalad-remake/datalad_remake/annexremotes/remake_remote.py

Lines 149 to 150 in 1f7b657

	# If no priority is configured, select the first instruction
	root_version, spec_name, this = next(iter(compute_instructions.values()))

[mslw]

I spoke to Joey and he confirmed that annex.geturls (which comes slightly earlier than the line above) returns URLs in a deterministic order - although that order might change when annex branch is merged.