Valence categories

[nautolycus]

Updated definitions relating to the bond valence model and added 'valence_units' to the units list.

Follows discussion in Issue #62

• I have added _valence_param.citation_id with some discussion of how this complements the VALENCE_REF category;
• I set the _enumeration.range for _geom_bond.valence to 0.0:10.0
• I haven't made any reference to the `9' convention for oxidation states that @vaitkus mentioned. I'm reluctant to hard-wire such a convention into a standard definition. I will contact the current maintainer of the valence parameters set to see if they are open to using '?' instead.

[vaitkus]

Looks good overall, left a few comments.

[vaitkus]

Thank you for the work, I am generally happy to merge this as is.

Would it be useful to leave issue #62 open for the time of being with the "Update specification" label as a reminder that [1] should be mentioned in the commentary chapter in Volume G?

[1] https://www.iucr.org/resources/data/datasets/bond-valence-parameters

[nautolycus]

Would it be useful to leave issue #62 open for the time of being

Yes, that would be fine - I've noted a number of issue sthat have been left open for this purpose, and will try to address those soon.