Improve rput then rpc performance

[Scott Baden]

Related to issue #140, which is more about semantics

Currently a stencil code implemented by Hadia is underperforming a solution that uses rpc_ff

We need to bring the performance of rput then rpc in line what that of the rpc_ff code variant

[Dan Bonachea]

This issue was triaged at the 2018-06-13 Pagoda meeting.

Scott noted he thinks this is important, and we will discuss it again soon.

Scott may adjust the milestone deliverables to get credit for this work

[Dan Bonachea]

Mass roll-over of unresolved issues to the next milestone.

[Dan Bonachea]

This issue was triaged at the 2019-07-24 Pagoda issue meeting and assigned a new milestone.

John agreed to try and produce an AMLong-based implementation of rput-then-rpc for the next release.

[Dan Bonachea]

This was resolved in pull request #110 at 0cff3e1

Issue #261 remains as a known performance defect.

Any future performance defects should be opened as new issues