SpeciesConstraints can forbid input species

[connie]

SpeciesConstraints only work on reaction families, not on seeds or libraries. Also they will forbid initial species from forming (such as Oxygen if maximum number of radical electrons is set to 1.) We should figure out a way to reconcile this.

1. Put all input species into an always allowed list as part of RMG class + Add a warning when parsing the input file declaring that your species constraints should be forbidding it and asking if you are sure.
2. Should we also allow things brought in from seeds and put them into the always allowed species list? Maybe not since often this includes a lot of species that RMG families can't deal with?

[rwest]

Your proposal yesterday to require users to explicitly ask for things to be permitted sounds like a good one. This could be in the same section where they specify the constraints, the specify exceptions to the constraints. If they include a starting species or seed molecule which is forbidden, then it stops with a clear explanation of the error.

[connie]

It might be cumbersome to manually allow all species from a seed. Probably in the speciesConstraints option itself we can have some flag that allows stuff in seed, input species, rxn libraries

[rwest]

Hopefully dealt with by #208 which we just merged.