Update simulation interface and add density to update_volume

[marjanalbooyeh]

This PR applies a couple of changes to the system and simulation classes:

• A simulation object can be instantiated in two ways now: 1) From a system 2) From an initial state and a list of forces
• run_update_volume in simulation now takes in a density parameter which basically calculates the final box length given the density and shrinks or expands based on that.
• target_box in system will be scaled by the reference length.
• A new method called calculate_box_length has been added to utils, which calculates the target box length given the mass and density. Both system and simulation classes call this method now.

P.S. In calculate_box_length method, I'm not sure if the mass conversion from amu to g is needed for all cases or not? This method is called in a few scenarios:
1- During system initialization where no scaling has happened yet
2- When run_update_volume is called from the simulation. This can be on a scaled or unscaled system.

Once we agree on these changes, I will update the unit tests and docstrings.

[chrisjonesBSU]

I left a couple small change requests.

[codecov]

Codecov Report

Merging #42 (332b46a) into main (4a626f9) will increase coverage by 1.40%.
The diff coverage is 97.10%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #42      +/-   ##
==========================================
+ Coverage   92.42%   93.83%   +1.40%     
==========================================
  Files          18       18              
  Lines        1321     1347      +26     
==========================================
+ Hits         1221     1264      +43     
+ Misses        100       83      -17     

Files Changed	Coverage Δ
hoomd_organics/base/system.py	89.51% <95.45%> (-0.28%)	⬇️
hoomd_organics/base/simulation.py	91.96% <97.29%> (+6.43%)	⬆️
hoomd_organics/utils/__init__.py	100.00% <100.00%> (ø)
hoomd_organics/utils/utils.py	100.00% <100.00%> (ø)

[chrisjonesBSU]

Since this PR makes quite a bit of changes to simulation.py maybe it would be a good time to use the validate_ref_value function in the ref value setters, just like we have in system.py?