explain "fast LOO" and "full LOO" in the docs, see

<#496 (comment)>

R/cv_varsel.R

@@ -22,12 +22,14 @@
 #'   performed. See also section "Note" below.
 #' @param nloo Only relevant if `cv_method = "LOO"` and `validate_search =
 #'   TRUE`. If `nloo > 0` is smaller than the number of all observations, full
-#'   LOO-CV is approximated by combining the fast LOO result for the selected
-#'   models and `nloo` leave-one-out searches using the difference estimator
-#'   with simple random sampling (SRS) without replacement (WOR) (Magnusson et
-#'   al., 2020). Smaller values lead to faster computation, but higher
-#'   uncertainty in the evaluation part. If `NULL`, all observations are used
-#'   (as by default).
+#'   LOO-CV (i.e., PSIS-LOO CV with `validate_search = TRUE` and with `nloo =
+#'   n` where `n` denotes the number of all observations) is approximated by
+#'   combining the fast (i.e., `validate_search = FALSE`) LOO result for the
+#'   selected models and `nloo` leave-one-out searches using the difference
+#'   estimator with simple random sampling (SRS) without replacement (WOR)
+#'   (Magnusson et al., 2020). Smaller `nloo` values lead to faster computation,
+#'   but higher uncertainty in the evaluation part. If `NULL`, all observations
+#'   are used (as by default).
 #' @param K Only relevant if `cv_method = "kfold"` and if `cvfits` is `NULL`
 #'   (which is the case for reference model objects created by
 #'   [get_refmodel.stanreg()] or [brms::get_refmodel.brmsfit()]). Number of