Enhance RunResults

[artgoldberg]

These to_hdf() calls in __init__() raise the warnings:

random_states_s.to_hdf(self._hdf_file(), 'random_states')
metadata_s.to_hdf(self._hdf_file(), 'metadata')

The warning is:

/usr/local/lib/python3.6/site-packages/pandas/core/generic.py:1996 PerformanceWarning:
your performance may suffer as PyTables will pickle object types that it cannot
map directly to c-types [inferred_type->mixed,key->values] [items->None]

[artgoldberg]

Also:

• Have hdf use disk more efficiently; for unknown reasons, in one example hdf uses 4.4M to store the data in 52 pickle files of 8K each, more than 10x less efficient than pickle.
• Speedup convert_checkpoints(), which runs slowly.
• Test all methods that take compartment id in a model with multiple compartments.

[artgoldberg]

did: make get_concentrations() work for all compartments

[artgoldberg]

master 9494ce9 did "Speedup convert_checkpoints(), which runs slowly"

[artgoldberg]

Finished, except for a couple of issues copied to #83.