Symmetry controls

[cjcscott]

We need to remove depreciated functions to add symmetry-related fragments, and add documentation to properly discuss how different symmetries should be added (see #95 (comment)).

Separately, we should also have a discussion about whether we want to continue to deal with symmetries in this way; the current approach may run into issues when deal with space group symmetries and a single approach should be possible.

[maxnus]

Hi Charlie,

for what it's worth, I'm planning to clean up the symmetry code at some point soon.
I didn't have any major changes in mind, except perhaps moving from chaining fragments (parent fragment->child fragments->grandchild fragments->...), where each inheritance step is characterized by a single symmetry operation only, to a setup where there are only parent and one level of child fragments, but their symmetry operations can be nested (say translation + rotation).

Separately, we should also have a discussion about whether we want to continue to deal with symmetries in this way; the current approach may run into issues when deal with space group symmetries and a single approach should be possible.

Can you elaborate what other approach you have in mind here?