@bob5 has multiple, similar posts about this topic, which is annoying. I previously posted in another of his threads: Flattening a 3-d helix - #16 by davidcockey
Due to the thickness of this part and the fact that it will be formed hot, there is no reason to be very concerned about starting with an exactly developable shape.
Why Gaussian curvature is not a good way to evaluate how close a surface is a developable shape: Verifying developable surfaces - #7 by davidcockey
Gaussian curvature and developability - #5 by davidcockey
Bob5’s part has a width to thickness ratio of 4:1. That means it when a phyical part it is bent within the elastic range it will deviate substantially from a developable shape due to Poisson’s ratio effects. The top and bottom surfaces will be anticlastic (double curvature in opposite directions). For the bent shape to be close to developable the width to thickness ratio needs to be much large such as a piece of sheet metal.
To greatly compound the deviation of Bob5’s part from a developable shape when fabricated the part will be made using a hot forming process. This means there will be large, inelastic deformation. The developable shape assumption is based on elastic deformation. With the hot forming there can be stretching, compression or both, and the final shape will depend on the details of the process.
Without expert knowledge of the forming process and more detailed analysis using the results of UnrollSrf or UnrollSrfUV on the not exactly developable surface may be a reasonable starting point. But I would consult an expert on the forming process before investing much time or money in trying to optimize the bland to be cut by waterjet.
After the hot forming the holes will be distorted considerable from cylindrical.
(Update - This is incorrect.) A developable surface is possible with the given edges.