While I appreciate the sentiment, it really isn't possible or even desired. What we know about genetics is like the volume of water in the Great Lakes, and what we don't know is comparable to the volume of water in the world's oceans. While it seems like we know a lot, we really know very little, almost nothing in comparison to what we don't know.
Take sickle cell anemia. The disease is a recessive trait, and both parents have to be a carrier in order for the disease to occur. It is believed that carrying one gene for the trait improved resistance to certain parasites, and thus an evolutionary advantage, allowing the gene to spread. To eradicate the disease with crispr, you'd have to eradicate it from all of the carriers, which could have the potential for negative consequences.
In addition, in order to eradicate all genetic diseases, you'd have to genetically test everyone, even those who don't want to be tested because they are reasonably concerned about their privacy. Moreover, what if a sickle cell (or other disease) carrier, who is perfectly healthy, doesn't want to have their genome edited? Do we force it on them, or just sterilize them so that they cannot breed? Obviously there are some serious ethical considerations.
Further, crispr is not perfect. There can be off-target effects. The use of it may be warranted when an infant is going to die anyway, but what about scenarios where the issue isn't fatal? There is always a risk of introducing an unintended genetic defect, and widespread use greatly increases that risk.
Again, reducing devastating disease is a laudable goal, but we're just hairless apes tinkering around with the building blocks of life, and don't know near enough to eradicate genetic disease.