Rewrite bonded forces routines in C

[Lun4m]

This PR potentially closes #11. Here's the checklist:

• [x] compute_bond_forces.c
• [ ] compute_angle_forces.c, should be straightforward.
• [ ] compute_dihedral_forces.c, might not be that easy (need to implent linear algebra operations that are readily available in Fortran).
• [ ] dipole_reconstruction.c, same as for dihedrals.
• [ ] Import methods? As of right now I haven't wrapped the C functions into PyObjects, which would allow the usual from force_kernels import ... because it kinda looks tedious to do (especially also considering that we are using NumPy arrays and it might not be trivial).

Note: this PR needs careful review since it's the first time I write in C.