I have some reasonable experience designing bowtie antennas and transmission lines, also in CST (though it's been a while since I've used CST specifically).
What it looks like is that the feed impedance for the coplanar stripline you're using to feed the antenna is too high. The impedance graph you're showing appears to be just the line impedance for the coplanar stripline feed, and not the antenna itself. This is calculated at the waveguide port, and is governed by the stripline geometry.
Probably a better way to design this is to break up the design into 3 steps.
Simulate and design the bowtie antenna, with the port defined at the bowtie elements. This will likely be a lumped port. Then modify the bowtie geometry as necessary to get your desired center frequency, bandwidth, and impedance.
Simulate and design the feed network. Most likely, you'll have to make the gap between your copper traces smaller to make the feed network also match 50 ohms.
Create a new simulation with the entire structure based on the previous two steps, to make sure it works together.
Also, the bowtie design calculator you linked has no details about the substrate. Based on this, you can only really use it as a starting point for your own design, since it's not going to be accurate for a bowtie on FR-4 substrate.
