[release/v1] Include more information on the role of "units"

[mkavulich]

Expanding on discussion here, it was discussed at a dedicated meeting that we should clarify the role that "units" play in the Standard Names, that they function as "preferred units" (SI when possible) that also indicate the dimensionality of the variable.

Also, should units be included for all standard_name entries, including things like chemical species?

[mkavulich]

Potentially related to #63

[mkavulich]

We should consider a new xml entry "base_name" or similar to not have non-standard-names listed as tags

[grantfirl]

This may be relevant: https://cfconventions.org/cf-conventions/cf-conventions.html#units

[grantfirl]

I agree that we need to be clear that units in the standard name dictionary DOES NOT mean that wherever a variable appears with a given standard name that it can be assumed to have that unit. One still needs to pay attention to the units metadata where the variable is used and make sure that the metadata corresponds to how the variable is used in the code. This is the main reason why units exist in the metadata!

[mkavulich]

Resolved with #129