The  rationale is  that the  Algo.spid that  gave processes  access to
their pid via their neighbors was a bit weird. Also, it makes the anonymity
of the algorithm more explicit: an access to the pid could appear anywhere
in the program, while now, the pid must be in state.t!

Now, if the  user needs the pid,  he must store it  explicitly in the
state via the initialization function.

To force that, I've changed the Register interface (and thus GenRegister)

The rationale is that:
- in lurette mode, they should be the same for all algos already
- in batch mode, action declarations are ignored anyway (except in custom mode)

Indeed,  the  generated  oracle  and daemons  were  not  syntactically
correct in such cases, e.g., if an action was named "I(p)".

Doc: Add some odoc generated html files to the rdbg section of the user guide

- do not hard-code the size of the graph in the ml files
- do not print the whole state of big graphs. Indeed, for graphs of
  10000 nodes, if we print the path (which is itself an array of size 10000)
  we obtain 100 000 000 variables, which is too much for the rif file.

(i.e., fold_left instead of map + flatten + @)

Indeed, for graphs of 10000 nodes on the dfs, we need to operate
on list of 100 000 000 variables, which overflow the stack.

to be consistent with  the behavior of lv6 and lutin  (and work with the
lurette stdin.stdout mode), and so that lurette works with sasa !

as it is where they belong.

Indeed, during the explanations, it makes more sense to start with the state.

Doc: rework the documentation of the Algo API.

Its basename is the basename of the name of the .dot file.

Also provide a =--gen-register= option generates such a registring ml
file. --gen-register also generate a state.ml skeleton file.

The Makefile test/Makefile.inc also suppose that the types of the state
is defined in state.ml

To do that, I've reorganized the code

Test: with ocaml 4.08, it is impossible to load the same function twice.
Hence the need to define p.mli files for graphs with several nodes.

Also, track missing files and the alea-coloring exemples.

- the move_count node was wrong (missing parenthesis)
- the lemma an the theorem only holds for locally central unfait demon

Moved the python environment into 'tools', added Project object in the python environment alongside minor changes, created some scripts to test the environment, and changed make_dot function in topology.ml such that it removes the path of a file before using it as graph name.

Added tools/test/skeleton to test gg and gg-deco. Made also small modifications to make_dot (in topology.ml)

The generation scheme is now makes it easier to write oracles that do
not depend on the topology.  The idea is to generate an oracle
(specific to the graph) that includes a file that the user should
provide, and which ougth to define a generic oracle, i.e., depending
on m and n (process and actions numbers).