@Jason Griffin good question, and honestly one that comes up a lot when people start pushing the module into real power-constrained deployments. I don't have a precise published TX current figure to hand you right now — that level of detail isn't in the public docs yet, and I'd rather tell you that than give you a number that sends your power budget sideways.
What I'd recommend in the meantime is just measuring it directly at the bench before you lock the enclosure — throw a current probe or a low-side shunt on the rail during a TX burst and scope the peak. With the supercap system's low ESR you'll actually get a pretty clean picture of what the voltage dip looks like under that transient load, which is more useful for your sizing calc than a datasheet number anyway since you'll see the real behavior with your specific cap sizing and trace impedance.
On the TX power control question — that's definitely worth flagging directly to Moamen through the Dev Hub or the Command Center, because if the STM32WB07 core supports adjustable TX power levels via the firmware (and it likely does at the radio layer), exposing that as a configurable parameter would be genuinely useful for exactly this kind of use case. Running at max output for a 200m link is a real waste and it's the kind of firmware feature that would get a lot of traction from the community.
For the GNSS + BLE simultaneous draw situation, I'd design around worst-case both-active peak and then let the actual bench measurement tell you whether you need extra cap headroom or if duty cycling the GNSS fixes is enough to keep the rail stable.