One of the crazy things about specifying a bleed resistor is that it is based on things like heat dissipation at the time it is not doing anything useful, just sitting there wasting energy. The amount of stored energy that it actually needs to bleed if the input voltage fails or is switched off is very small. I chose to forget it altogether - I didn't want all that extra heat and wasted energy when, in practice, the drivers do the job very effectively. The only time a bleed resistor would be useful is under unusual conditions, during testing, for example. I'm happy to be aware of the issue at such times - it's no worse than avoiding the mains connections, for example. I'm not criticising people who do fit them, but for me it falls into the same category as a "cabinet open" switch. Commercially you should fit one of these in case some clown opens the cabinet and puts a finger where they shouldn't. For home workshop use, I want to be able to run with the door open sometimes - for example, to measure the power supply voltage to see what happens when it is turned off!

I have three e-stop switches around the machine and limit switches on all axes (except lowest position, of course). For me, that is a reasonable set of safety features; there's an MCB where the power enters the cabinet, a safety relay, and a fuse on the motor PSU (because it was fitted when I bought it). The only failure I have had on my machine was that fuse failing - through age, it seems, as there was no overload. Other people fit MCBs, fuses, etc, all over the place. Opinions vary!