Issue 3518: Support temporary divisi staves

Issue 3518: Support temporary divisi staves

This provides the low-level support for temporary divisi staves by
adding a `VerticalAxisGroup.remove-layer' property of type integer that
interacts with the "Keep_alive_together_engraver": when set to a numeric
value, staves with the same numeric value are kept alive together as one
group.  Of several such groups with live staves, only the one with the
lowest common numeric `remove-layer' is retained.


Also contains commits:

Regtest for VerticalAxisGroup.remove-layer (divisi staves)

Reformat define-grob-properties.scm to avoid column 0 parens in strings


This is actually a more low-level interface than the discussion in the
issue is mostly about.  Since the provided mechanisms are considerably
simpler and more flexible than envisioned in the discussion, it makes
sense to let people integrate this into their existing
frameworks/workflows without prescribing particular wrappers at this
point of time.

[lemzwerg, 2014-08-20 14:48:41]

LGTM.  Very nice!  I guess it makes sense to actually provide `\boring' and
`\tricky' with better names...

[dak, 2014-08-20 15:01:22]

On 2014/08/20 14:48:41, lemzwerg wrote:
> LGTM.  Very nice!  I guess it makes sense to actually provide `\boring' and
> `\tricky' with better names...

For a real use case/interface (which would also warrant ending in a snippet
and/or the documentation), \boring and \tricky in the current form would not
really work well: the whole point of a divisi passage is to disentangle a
situation that is too hard to show properly in a single staff.  But that means
that we want the passages now placed between \tricky and \boring to be replaced
by spacer rests or similar in the single staff.  Otherwise we'll get NoteColumn
collision warnings and similar for the tricky passages and, worse, have the
overcrowded tricky single-staff passages (which will not appear on paper)
determine the horizontal spacing in the divisi passages.  Of course, some
interference of ultimately disappearing staves cannot be avoided, but at least
those passages where the complexity is known to _demand_ the divisi staves,
cramming the material into a non-divisi phantom staff should not be allowed to
determine warnings and spacing.