Validating the number of samples

[wasade]

Improvement Description
A user on the QIIME 2 forum encountered a memory error when running a FeatureTable containing a single sample through core-metrics-phylogenetic. The error stemmed from Striped UniFrac not appropriately checking for a single sample (noted in biocore/unifrac#34). While the actual error stemmed from underlying code, we may want to consider adding in a helpful message to the user that this action is not relevant for a single sample.

Current Behavior
When a FeatureTable with a single sample is provided, and error is produced.

Proposed Behavior
Error, and provide a polite message to the user indicating that beta metrics are only applicable to tables with 2 or more samples.