[[Imported from SVN: r6295]]

[[Imported from SVN: r6294]]

[[Imported from SVN: r6293]]

[[Imported from SVN: r6231]]

[[Imported from SVN: r6226]]

[[Imported from SVN: r6215]]

That makes the code more readable, which is more important to me than
speed.  The stuff is difficult enough as it is.

[[Imported from SVN: r6214]]

[[Imported from SVN: r6213]]

factor out the computation of the linear part of the reference-simplex-to-standard-simplex computation

[[Imported from SVN: r6212]]

[[Imported from SVN: r6211]]

[[Imported from SVN: r6210]]

[[Imported from SVN: r6209]]

[[Imported from SVN: r6046]]

[[Imported from SVN: r5996]]

[[Imported from SVN: r5995]]

[[Imported from SVN: r5994]]

[[Imported from SVN: r5993]]

[[Imported from SVN: r5992]]

[[Imported from SVN: r5991]]

[[Imported from SVN: r5989]]

[[Imported from SVN: r5988]]

[[Imported from SVN: r5986]]

Compute the local corrections in local coordinates on the tangent bundle now, instead of using the embedding in the surrounding space

[[Imported from SVN: r5985]]

[[Imported from SVN: r5984]]

[[Imported from SVN: r5966]]

[[Imported from SVN: r5965]]

[[Imported from SVN: r5964]]

For this now the TangentVector class of S^n has only n components,
while the EmbeddedTangentVector class still has n+1.
The code around it needs to be fixed accordingly.

[[Imported from SVN: r5963]]

[[Imported from SVN: r5962]]

out-comment the use of method computeGeodesicDifference.  It currently doesn't work for all target spaces

[[Imported from SVN: r5959]]

[[Imported from SVN: r5958]]

remove the possibility to solve the inner problem with a truncated cg.  It never really worked and complicates the code

[[Imported from SVN: r5951]]

[[Imported from SVN: r5947]]

[[Imported from SVN: r5946]]

[[Imported from SVN: r5941]]

[[Imported from SVN: r5940]]

[[Imported from SVN: r5939]]

[[Imported from SVN: r5938]]

[[Imported from SVN: r5937]]

[[Imported from SVN: r5936]]