[DOC] - Approach for new benchmark addition

[sbradnam]

What where you looking for

Method to implement a new benchmark. Information on adding benchmarks now that input models are hosted separately form the main repository.

Is the information available in the doc but not where you would expect?

No.

Is information completely lacking or partial?

Information is partial, changes to the process have occurred since the documentation was updated.

Where would you insert the missing information?

In the documentation.

Additional context

No.