[BUG] PDF calculation fails for trajectories containing dummy atoms.

[MBartkowiakSTFC]

Description of the error
Normally, the Pair Distribution Function works fine, but it seems to fail at least for one trajectory I tried.

Describe the expected result
PDF should always work.

Describe the actual result
For a DL_POLY water trajectory (TIP4P), it failed instantly with the error:

File "MDANSE/Framework/Jobs/DistanceHistogram.py", line 184, in run_step
van_hove.van_hove_distinct(
File "Extensions/van_hove.pyx", line 110, in MDANSE.Extensions.van_hove.van_hove_distinct
IndexError: Out of bounds on buffer access (axis 0)

Suggested fix
Most likely the problem is caused by dummy atoms in the system.

Additional details
I could get the correct result by removing the dummy atoms using TrajectoryEditor.
Excluding the dummy atoms using AtomSelection was not enough.