I've never seen anything to suggest that tasting was involved in the selection of slaves. That said, there were many, many slave markets and each had its own peculiarities.

In his masterful Soul by Soul, Walter Johnson describes many of the practices slave buyers and sellers used when selecting a potential slave. Skin color, body shape, behavior, eye movement, and a host of other characteristics were weighed for what was called 'likeliness.' If a potential slave was deemed 'likely' s/he would do well in the role the slave owner chose.

The key to Johnson's argument (well, one of many arguments) is that what made a potential slave 'likely' varied from one purchaser to another. Some wanted the larger men to dominate, inflating the owner's sense of power. Some wanted a woman whose skin color was lighter and who had 'white' features, perhaps acting out some sort of semi-sexual dominance. There was a sort of projection from buyer to slave and vice-versa. The slave market became a bit of performance theater in the worst possible way.

Slave traders became adept at promoting various qualities of their potential slaves to different buyers. Setting aside the vileness of human chattels, it's not much different than a salesperson/purchaser arrangement today. Salespeople read the buyer to know what s/he wants and then promotes various qualities of their product to make a sale. There are numerous examples in Johnson's book, and these are but a few.

As to tasting? Who knows. Maybe some huckster said that you could taste disease or diet or whatever. I've never seen evidence of this, but would not surprise me in the least.

If his interests you, the chapter in Johnson's book that's most closely related to your question is called "Reading Bodies and Marking Race." Truly good stuff.