The calculators do not do a good job of distinguishing genotype (identity of the genes) from phenotype (appearance).
Definitions:
Homozygous = the two genes in a gene pair are the same. Two copies of the normal gene or two copies of a mutant gene.
Heterozygous (het for short) = the two genes in a gene pair are NOT the same. A normal gene and a mutant gene or two different mutant genes.
Genotype clown and normal = het clown --> normal phenotype (clown gene is recessive to normal gene)
Genotype fire and normal = het fire --> fire phenotype (fire gene is codominant to normal gene)
Genotype spider and normal = het spider --> spider phenotype (spider gene is either dominant or codominant to normal gene; jury still out)
Genotype pastel and normal = het pastel --> pastel phenotype (pastel gene is codominant to normal gene)
Genotype pastel and normal, spider and normal = het pastel, het spider --> bumblebee phenotype
Herpers generally reserve "het (mutant gene)" for a normal-looking snake with a gene pair made up of a recessive mutant gene and a normal gene. This is a holdover from the days when the only known mutant genes were recessive to their normal gene counterparts. Those days are long over, and herpers need to adjust.
As a het clown royal python looks normal, the clown gene is recessive to its normal counterpart.
A spider royal python has a spider mutant gene paired with a normal gene, giving the spider (phenotype) royal python a het spider genotype. As a het spider royal python does not look normal, the spider mutant gene is not recessive to its normal gene counterpart. Clear as mud?
The calculator is assuming that the user knows all this. And knows that a spider het clown royal python looks like a spider royal python. Because the het clown part of the genotype does not change the phenotype.
Sorry if people think I'm belaboring the point, but you did ask to understand hets.