Merge pull request #3949 from RosettaCommons/roccomoretti/shorten_sewing_runs
Shorten legacy sewing runs.
A quick & dirty profile of the sewing integration tests indicates that a substantial amount of time is being spent in Model::model_end() calculations in the Hasher::transform_model() function. This is likely superflous recalculation, as it's only being used as a loop end condition.
Fix the code to only calculate the end iterator once. It doesn't solve the integration test timeout issues, but it should reduce the runtime somewhat.