Do this by making RodAssembler more similar to the generic
GeodesicFEAssembler: the constructor argument is now a
general LocalGeodesicFEStiffness, and not something particular
to rod problems.

Throw out the AmiraMesh support instead.  I have not seen
anybody using Amira for years.

[[Imported from SVN: r9982]]

[[Imported from SVN: r9616]]

[[Imported from SVN: r9614]]

[[Imported from SVN: r9579]]

[[Imported from SVN: r8206]]

[[Imported from SVN: r7614]]

[[Imported from SVN: r7028]]

[[Imported from SVN: r5938]]

[[Imported from SVN: r5915]]

[[Imported from SVN: r5723]]

[[Imported from SVN: r5347]]

[[Imported from SVN: r5099]]

[[Imported from SVN: r4177]]

[[Imported from SVN: r4125]]

[[Imported from SVN: r4124]]

[[Imported from SVN: r4079]]

[[Imported from SVN: r4056]]

RodSolver is now called RiemannianTrustRegionSolver, and can be parametrized with the TargetSpace.  Some remains of RigidBodyMotion are still there, though.

[[Imported from SVN: r4041]]

[[Imported from SVN: r4027]]

now works on a per-element basis and has been moved to the local rod assembler.
That way, it has lost most of its 1d-ness.  In the future, it will be usable
for other problems as well.

[[Imported from SVN: r4019]]

[[Imported from SVN: r4018]]

The class 'Configuration' is now called 'RigidBodyMotion', and can be parametrized by the space dimension

[[Imported from SVN: r4003]]

[[Imported from SVN: r3990]]

[[Imported from SVN: r3489]]

[[Imported from SVN: r3286]]

[[Imported from SVN: r2929]]

[[Imported from SVN: r2618]]

[[Imported from SVN: r2524]]

[[Imported from SVN: r2020]]

[[Imported from SVN: r1717]]

[[Imported from SVN: r1680]]

[[Imported from SVN: r1634]]

[[Imported from SVN: r1633]]