Write roman ts and repeats

[MarkGotham]

Write complement to #1378 along with a bit of clean up

[coveralls]

Coverage decreased (-0.0003%) to 93.036% when pulling a25cdf5 on MarkGotham:writeRoman_ts_and_reapeats into 5417b3c on cuthbertLab:master.

[mscuthbert]

Please see comments on removing tested/documented code without replacements.

[MarkGotham]

Ok, noted.

The mass deletions are an attempt at simpler solutions and strike me as a good fit for the contributing docs' invitation to remove 30+ lines when 1 will do nicely. All the same, while I have misgiving about rnString (for that reason among others), I'll go ahead and reinstate it with the minimum changes to accommodate new tokens (i.e., repeat marks).

For reference, the process is about the addition of specific tokens in a specific order.

• measure (required)
• (optional) start repeat
• RNs (with beat except for b1, at least one required)
• (optional) end repeat

[MarkGotham]

This PR has been superseded by another, subsequent one for the same issue at #1435.

I wonder if there's an integration action for checking new PRs against those in progress that would prevent this?

In any case, I've been through to pick them apart and what remains as new here is simply:

• adding a doc test and
• one case of specifying insert() at position 0 (as requested)
• misc clean up and clarification

A small contributions for sure, but a positive one, and makes this (1393) ready for closure.

Let me know if you spot any issues / anything missing.

[mscuthbert]

The contributions here are enough to still merge after the changes requested are made.

If you know of an easy tool that can prevent duplicated work, happy to add it.

[mscuthbert]

LGTM. Thanks!