Your gage remains at the top because generally speaking your filler neck is a couple inches in diameter and can hold an extra gallon or two. On larger vehicles, like trucks, it can hold up to 4 gallons extra depending on how it is routed. The float in the tank can only measure the tank itself. Therefore you may get an extra 10 - 20% at the top because of the filler neck. Thus the gas gage stays at the top for what can seem to be a decent amount of time.
In addition, most of the float systems don't reach the top of the tank. This is simply design constraints and has NOTHING to do with trying to 'fool' people.
While a lot of what you say makes sense, I sincerely doubt that the filler neck on the average sedan can hold, "a gallon or two". One gallon is 231 cubic inches. If the filler neck has a 2 inch inner diameter, it would have to be over 18 inches long to hold just one extra gallon. I'm struggling to imagine what sort of convoluted routing would result in a filler neck that was over 6 foot long and thus capable of storing those "extra 4 gallons" that you claim some trucks can manage.
In fact, measuring both my sedan and coupe gives a length of about 10 inches for the filler tube on each, with the rear tank of the F-150 having a filler length of about twice that. That means the cars have maybe an extra half-gallon and the truck an extra gallon if you fill until fuel is coming out the filler, which pretty much nobody ever does. Also, the F-150 filler tube pretty clearly isn't more than a 2" inner diameter and that's probably being generous. There just isn't any way that a system would hold 4 gallons of gasoline in the filler neck.
The fact is that most gauges are non-linear today. Gauges for things like temperature or oil pressure are often programmed to stay in the middle of the display range unless there is a large change. This maintains the usefulness of the gauge as a warning indicator but reduces the number of morons who come in and complain that their car runs 5 degrees hotter than their neighbor's and demand that it be fixed despite it being a non-issue. Sometimes, even the markings on a gauge make this nonlinearity clear.
For example on the current model of Toyota Camry (a car that I have driven extensively as a rental over about 9 months and probably 5,000 miles last year), there are 15 evenly-spaced markings between the Full and Empty marks. Seven markings between Full and half, and 7 between half and empty. However, there are only two marks between the full and 3/4 marking, and there are similarly only two markings between the 1/4 and empty marking. (
See Photo.)
That gives increased resolution around the middle of the tank, which makes sense (and may even be a technological limitation as you mention), but it's clearly calibrated to be nonlinear. Recording mileage makes that quite clear.
In contrast, my 951 and my S70 both have linear fuel gauges. In fact, the Volvo's gauge, though analog is marked in terms of gallons remaining and won't reach "full" (18 gallons) unless I top off the tank to near-overflowing. Perhaps this is simply differing philosophies, but it doesn't seem to be difficult to get a float that goes all the way to the top of the tank and it's very clear that many gauges will read "Full" even when the tank is a gallon shy of published capacity.
ZV