Allow `filter_taxa` to subset anything that has names, length, and can be subset

[zachary-foster]

I was about to add the ability to work with DNAbin from the ape package, but thought that a more generic solution might be better. Anything could be subset by filter_taxa if it:

1. Is one-dimension, with a known length (i.e. works with length function)
2. Can be subset (i.e. [ changes the value of length).
3. Has per-element names that can be set (i.e. works with names and names<- function)

If I can write a function to test if these are true, then many list-like things could be included without a lot of custom code.