Dark matter and dark energy proposals: maintaining cosmology as a true science?

I consider the relation of explanations for the observed data to testability in the following contexts: observational and experimental detection of dark matter; observational and experimental detection of dark energy or a cosmological constant $\Lambda$; observational or experimental testing of the multiverse proposal to explain a small non-zero value of $\Lambda$; and observational testing of the possibility of large scale spatial inhomogeneity with zero $\Lambda$.