[makedoc]

[almost]

[almost]

[almost]

[almost]

[almost]

[almost]

This test is failing on travis-ci with some
particular debug settings.

@gtribello suggested to remove it and add it back on master branch.
This is motivated by the fact that people using these features should
already be working on the master branch, where this functionality has
been improved and apparently does not fail.

[makedoc]

[makedoc]

[makedoc]

[makedoc]

[makedoc]

[makedoc]

Related to #165

[makedoc]

[makedoc]

[makedoc]

I had to tweak a but the new regtest (rt-mpi9) so as to make
it compatible with new filenaming for multiple replicas
(e.g. plumed.dat.0 -> plumed.0.dat)

Related to #168

This bug was introduced by me in 5e3aaf3a
The barrier on multi replicas was due to overcaution and
is only necessary when using WALKERS_MPI

The kernels were normalised incorrectly and their derivative when the
covariance matrix was non-diagonal was computed wrongly.  This check
affects a number of regtests.  analysis/rt0 changes beecause the kernels
were normalised incorrectly.  basic/rt44, basic/rt48, basic/rt49, basic/rt52
change because the derivatives output in the fes file have changed sign

This is to avoid using an outdated file

This hack was necessary to support DESTDIR.

Now that all installed files are placed in src/lib/install,
this is not necessary anymore