If a planet lay in the Goldilocks zone but doesn't support life, why is that? Lack of water? No magnetosphere? Too many asteroid impacts? Metal poor? If life isn't there already, the conditions may make it hostile to permanent settlement. If life is present, then at least something was able to adapt to the conditions. If you have significant amounts of liquid water, good partial pressure of O2, a water cycle and a functioning magnetosphere, then you're already a pretty big leg up on all the other planets we currently know about. That's a heck of a lot of terraforming already done for you, possibly some or much of it due to your new organic friends.
We don't know what life could look like elsewhere in the universe. But we do know a few things: the laws of chemistry and thermodynamics appear to hold for other solar systems, to the extent that we can infer them (from spectroscopy, redshift, etc.). Since life is an information-dense chemical system, we know that it requires fairly sophisticated molecules to achieve its effects. While simple crystals "grow", they don't do so in the kind of entropy-lowering way that cells do. That means it will need molecules that can become large and differentiated.
Carbon is a flexible molecule because of its +/-4 oxidation state. Silicon is in the same group, and should have roughly the same flexibility. But the difference in atomic weight causes the chemistry to differ significantly. So while we can make silicon polymers, we don't know how to make silicon amino acids/enzymes, and have reasons to believe that silicon is not capable of supporting these more elaborate molecules. Of course, the rest of the Group 14 elements (Germanium, Tin, Lead) are hopelessly useless for biology.
So if we have convinced ourselves that life is actually pretty likely to be carbon-based, why should we assume that it's anywhere compatible with Terran biology? Well, for starters, it turns out that our genetic code is highly optimized, from an information-theoretical perspective. So while you could have an equivalent code with a different assignment of codons to amino acids, the 3-base codons, 4 base DNA/RNA code appears to be at least a local optimum. For this reason, other biological systems are likely to have at least a similar configuration. Of course, the chirality choice appears to be free, so you could get unlucky and run into life with incompatible chirality.
But we know that amino acids can be created spontaneously under adequate conditions, and thus seem likely to form a good building block on other planets. We have detected amino acids in space. Yet another reason to believe that exobiology is likely to not only be carbon-based, but amino acid-based. RNA and DNA are not a terribly big step up from amino acids themselves, so they represent one of the simplest viable molecular information encodings.
So if the exobiology produces a similar/compatible set of amino acids, then it is possible that your colonists can eat them. After all, digestive enzymes can't and don't target particular proteins and peptides, but rather just particular amino acid bond types. If earth animals can also eat the local flora and fauna, then the colonists should be able to eat a mix of both local- and xeno-food. Terran plants will likely not be able to participate in the local symbiotic networks, and may compete unfavorably with the local flora. But if the other conditions are suitable (light, water, temp), then the colonists should be able to grow Terran plants in isolated greenhouses.
We like to hedge our bets and cover our bases and assume that life can take on some mysterious form that we cannot even imagine. But the reality is that chemistry and thermodynamics significantly restrict the kinds of magic that any plausible xenobiology could employ. I am willing to go out on a limb and claim that if life exists elsewhere, there is more than 50% chance it is carbon based, and that if it is carbon-based, there is more than 90% chance it uses amino acids, and that if it uses amino acids, that there is more than 90% chance that it uses some to many of the same amino acids that we do.
Note that there is some tolerance in amino acid utilization. If a protein calls for one amino acid in a particular location, but ends up with a different but similar one instead, it can still be mostly functional. Obviously, this is not accidental. When we say that the Terran genetic code is universal and "optimal", we mean that it is maximally robust to the most frequent kinds of noise/errors in transcription because similar codons are assigned to the same amino acid, and codons for different amino acids that are likely to be mistaken for one another code for functionally similar amino acids. Thus, even if xenobiology uses a different but overlapping amino acid set, there is a possibility that some of the proteins used by xeno-life is still partially compatible with Terran life.
Writers often cast biology as a binary compatible or incompatible, but the scale is much fuzzier than that. Even on earth, members of different species are supposed to be infertile together, and yet we have regular inter-species hybrids like mules and ligers. Such hybrids are not infinitely far-fetched between Terran and xeno-species, though the success rate is likely proportional to how large/complex the individuals are (and thus, most likely limited to unicellular/bacterial colonies).