Is it by design that gnomAD variants are not called for human genome?

[davidyuyuan]

Is it by design that gnomAD variants are not called with mpileup and call? For example rs3064744 is *28, *36 & *37 in UGT1A1 in Pharmacogenomics. I'd like to mention that this variant is in the reference genome GRCh38.

I just want to understand the behaviour of bcftools whether it calls all known variants with RS ID defined in human genomes or it just calls the variants not part of the reference genome.