I'm honestly not that familiar with Dash, so I can't comment on that aspect.
You state that the internal server will do the heavy lifting. In other words the python environment will be running as a process on that server and you will connect to it remotely from your laptop. In that case you are right and wouldn't need that much processing power.
However, you boss must have a reason (why not ask him?).
A big question is whether this internal server of yours is a production server. If so then I think your boss's expectation may be that you will use your laptop as a development environment and then deploy to the server so that your workloads don't affect other users negatively. This is because development work can get quite resource intensive as you iteratively experiment, tweak and test before settling on a solution. This is particularly true if you need to work with large data sets and complex workflows.
How do the (I assume) Lenovo P15 specs compare to the server?
What kind of visualization will you be doing? Are we talking about simple ones or complex things like 3d animations? The biggest selling points of the P15 are the Xeon CPU (for CPU intensive things) and the professional grade Quadro RTX GPU. These machines are often sold in the CAD, engineering and 3D graphics markets. Do you need this? Servers typically don't come with powerful GPUs.
I'm also wondering why you would object to getting a P15, if it isn't your money being spent. Reasons may include that it is big and heavy and runs hot or you don't want to be responsible for such an expensive piece of kit or because, powerful as they are, workstation class laptops are not good at certain things (like gaming).