Ideally, you’d test at least 2 or 3 lights of the exact same model, to be sure you’re not either getting a “dud” or an “exceptional” light. Maybe that’s what you mean when you say 3-5 flashlights. But, even if you just test 1 unit each of 5 different models, it’s informative.
Do the same test over and over and over again. It might take dozen of tries before you get a light to land in a particularly bad way, such as on the lens or switch where it breaks.
If you don’t care about breaking the light, increase the severity of the testing, to figure out just when the light will break. e.g, Test 1 meter drops, then 2 meters, then 3 meters, etc.
For drop tests, do it onto concrete or hard rock. Asphalt is just too soft to be a good drop-test.
Water tests, might include some tests for using the switch while under water.
Personally, don’t spend too much time on hot/cold tests. You can do it, but any light that can’t operate within –20C to +60C is junk. So, no need to test it more than once for each light.
In addition to drop tests, do “throw tests”. This would simulate dropping the light while cycling, or out a car window. Maybe do it on a bumpy surface, to give the light lots of skips and bounces. You could also do this by simply tossing the light out a car window while driving.
The switch is often the weak point, so test that somehow.
Water tests might include salt water. That’s often much more destructive, especially for long periods of time.