All CondSCVI models for spatial deconvolution yielding per spot cell-type proportion and cell-type activation state.