I finally upload the GREX stuff. It seems to work, but I did not really
debug it (there might be wrong units or so). I also provide a patch
for gromacs 4.5.3. Interface is still a bit tricky, and probably it
will be necessary to adapt it a bit as we try to port it to other codes.

The overall flow is:
* at steps corresponding to attempts, positions/total energy are
  saved on a temporary buffer when normal plumed is called.
* when repl_ex is called in gromacs (which, in the case of gromacs
  is after moving atoms, so with different positions) and
  the partner replica is decided by gromacs, a fake exchange of all coordinates
  is done inside plumed and a new calculation is also done.
  Notice that since presently all "analysis" tools are acting in the
  calculate() method, they are called once more at this stage. This
  should be fixed.
* Then, change in local bias potential and foreign bias potential
  are collected by gromacs and added to the acceptance.

Notice that with this approach the plumed.dat files used for different
replicas could be completely different (but for UNITS! this probably
can be fixed, even if it seems a bit stupid to me), and, a part from
trivial errors which could be there, the acceptance
is evaluated consistently

I moved the tools to open a file with a specific prefix to PlumedMain.
This allows easily to use them also to open the input file, which in
turn means that we will be able to do bias-exchange and Hamiltonian
replica exchange

This is useful to compute acceptance.
I did it in the "dirty" way: if a value named "bias" exist inside any
action, it is accumulated into the total bias. I also changed the name
of values in Restraint and MovingRestraint so that the energy is now called
"bias" and non-capital letters are more consistent

It is now possible to automatically add a suffix to
all the files opened through the Action::fopen tool.
Opening is done as follows:
First the file+suffix is tried.
In case of error the file without suffix is tried.

In this manner, it should be possible to provide a single
input file also for replica exchange stuff. For output files,
it should make any difference.

I also added a stupid keyword to test this feature. Just use
_SET_SUFFIX pippo
in the plumed.dat file to add a suffix pippo to opened files

I am starting to merge my GREX branch, which should be
general enough to be used in any REX framework where
coordinates are swapped (a-la-gromacs). Since GREX
significantly modifies the MD-PLUMED interface through
the addition of several new cmd(), I tried to encapsulate
it as a separate objects with a cmd()-like interface.
Commands from MD starting with GREX are automatically
forwarded to that object. E.g.
plumed_cmd(plumed,"GREX init",NULL);
or so should initialize the GREX object, etc.

Up to this point I just upload the interface, later
on I will add the real GREX object, for which I still
have to do some cleaning.

In particular, I would like to be able to just do the
"forward" loop (calculate()) and not the "backward" loop
(apply()). In this manner, when computing foreign biases
for acceptance in replica exchange, I will only call the
forward loop, which computes colvars and biases. Clearly,
we should move printing/analysis/hills-adding stuff
outside the "calculate" method, so as to avoid doing these
things in the acceptance calculation. I wait for this further
cleaning since, until replica exchange is not actually there,
there is no practical difference. Also, we may decide to change
a bit the name of methods (calculate/apply are becoming a bit
misleading?)

Instead of deactivating actions at the end of the step, I prefer
to deactivate all of them just before activation. In this manner, all
dependencies are touched in the prepareDependencies method.

It is a very primitive "serialization" interface. I just put it here
so as to use it with replica exchange to send atoms to other replicas.
Still, we should plan something more general than this and replace it,
especially when we will need to pass between replica more complex objects.

This will be necessary for replica exchange

I also removed an "included" added by Davide a few commits ago. I think
it should not be here, but if it is necessary on some compiler we can
reinsert it. However, we should use "cstring" and not "string.h"

wrong behavior with mpi-plumed+non-mpi-md

I modified the "LOAD" command in such a way that if one
writes "LOAD pippo.cpp" in the plumed.dat file, pippo.cpp is automatically
compiled to a pippo.so which is then loaded.

A compile_options.sh file is generated by make in the src directory
and it is used to properly compile a .cpp into a .so (or .dylib on mac).
Just type "plumed mklib pippo.cpp". Still it should be improved in term
of linking of multiple .cpp files.

Modified the "plumed" script so as to simplify the addition of new
tools. To add a new tool just put a script in tools/scripts with extension
.sh. The script should understand the keyword "--description" which is used
to build a summary of available scripts in the main "plumed" script.

Groups are removed when GenericGroup objects are deleted

Dependencies are not generated when cleaning

Fixed a bug which was giving nan when distances with pbc were
computed in boxless systems.

There was a wrong call in the access to units

(valgrind)

Especially, I removed "0" atoms, which was always crashing.
I also put some example of groups and centers of mass

I also replaced math.h with cmath

Part of the interpretation of the cmd string can be moved here.
Also, it will be easy to define other classes with this interface
(I am planning to do that for gromacs-style replica exchange)

I prefer not to use strtok_r, which is not standard, not strtok, which
is not thread-safe.

Added some doc and removed declaration of an unused function

not using splines

file are still missing. And probably needs optimization (inline stuff?)

to having Grid and SparseGrid class and a flag dospline

cubic splines (from plumed1 at this time). At variance with Grid,
an additional vector<double> containing derivatives should be
provided when using the methods addValue and setValue.
The method getValue can be called with or without requesting derivatives.
The second class SparseGridWithSpline inherits from SparseGrid and
implements the splines in an analogous manner.

I forbid Actions to declare both an unnamed value and named values.
In a far future, the action name alone will imply passing the full structure.
Example:
d1: DISTANCE ATOMS=1,2 COMPONENTS
now only declared d1.x d1.y and d1.z, whereas d1 cannot be accessed
directly as a value (since it is a complex structure that we cannot handle
yet)