Should population_term and observation_term be mandatory in meta_forestly?

[yuliasidi]

Since each forest plot is specified for a particular population/subset of observations - do the population_term and observation_term really need to be defined? The actual subsetting is already specified in population_subset and observation_subset, therefore, requiring a user to give these subsets names for generating one plot seems to be redundant.

Please consider either removing these arguments or assigning some default values so users don't need to deal with the additional unnecessary parameters specification.

[elong0527]

Closed per #48