There's not really big difference in absolute terms of the risk of death w/ vs. w/o a vaccine. The difference is far less than 1% (say.. ~1 in 300 chance of dying without vaccine in a year vs 1:20,000 so 0.3% vs 0.005%). That does not mean vaccines are not effective. They are extremely effective! It just highlights the error in looking at differences in absolute risk, like 7% vs 8%.
The Bangaldesh RCT was great because it showed what masks do in a real world situation. Compliance was not great in the study but a decent effect was still seen! That suggests that masks are even more effective than the study shows assuming perfect compliance.
Your opinion is that an 11% decrease is not enough to justify use. Given the low cost of masking, most policy makers would disagree, especially in times of exponential growth.
The reason mask mandates are going away is because COVID incidence is rapidly receding in many parts of the US.
I don't pretend that a mask makes me magically 100% safe, but it's quite clear that they can help limit transmission with a relatively tiny burden on the user.