The way I see it is that so many people get by on nutrient-low, unbalanced diets that it's impossible to know exactly what the optimal combinations of food to eat really is. Humans are pretty resilient.
I doubt Soylent will be better or worse for you than the diet of a typical HN user if taken in the long term. Time will also tell: if after 5 years no adverse health effects are reported after a sufficient number of heavy users begin drinking it, then that's probably as best a "green light" as one could ever get.
There's still too much we don't know about how we get nutrients from food to be able to fully replace fresh unpackaged products with powders. I suspect we're still at least a few decades away from being able to do that safely on the long term.
Nutrition research has been terribly inconclusive for the past century and there's no indication that this is changing anytime soon. It's not the researchers' fault: the domain is incredibly complex when you take in consideration various timespans, all the various human generic permutations etc.