Because it is interesting to think about and would be an achievement if you could do it.
Interesting to think about, sure. But even if you could do it, it would be extremely expensive, which means nobody is going to try it just to see.
But let's think about this in a little more detail, since that seems to be the point. The easiest shape to do is a sphere, both because it's fairly straightforward to make (and calculate) and because it's going to be the most stable against hydrostatic pressure. So how much pressure are we talking about here? Well, let's play around a little with some numbers.
Let's see what it takes to make a 1 meter vacuum balloon. A 1 meter diameter sphere has a cross sectional area of 0.785 m^2. Atmospheric pressure is 101 kPa, or about 1x10^5 Newtons/m^2. If we divide our balloon into a top half and a bottom half, we can calculate the force that the two sides apply to each other as cross-section x pressure = 7.9x10^4 Newtons, which is a little under 9 tons of force. This will be applied along an area of 3.14 meters x thickness (assuming a thin shell).
Now, how thin would this shell need to be? For a 1 meter diameter sphere, we're displacing about 0.52 m^3 of volume, which is around 1.2 kg/m^3 near sealevel (good enough for the moment). So we're displacing about 0.63 kg of air, which is the upper limit on the weight of our sphere. The volume of shell is going to be
area x thickness = 3.14 m^2 x thickness, or
thickness = volume/3.14 m^2
Steel is about 7.8 g/cm^3 = 7.8x10^3 kg/m^3. If we made our sphere out of steel, the volume would need to be
mass/density = 0.63 kg / 7.8x10^3 kg/m^3 = 8x10^-5 m^3,
so the thickness would be
8x10^-5 m^3/3.14 m^2 = 2.5x10^-5 meters
or 0.025 mm thick (clearly not enough to support that much pressure)
Of course, we could build it out of something stronger, and also possibly less dense. As one example, we'll consider diamond, with a density of about 3.5x10^3 kg/m^3. Run through the same calculations and you'll get a thickness of about 5.6x10^-5 meters.
So let's suppose we went with diamond. The cross-sectional surface area of our shell is now 3.14 m x 5.6x10^-5 m = 1.76x10^-4 m^2. The force on the two halves of the shell is 7.9x10^4 Newtons, so the pressure is 7.9x10^4 / 1.76x10^-4 m^2 = 4.5x10^8 Pascal.
The Young's modulus for diamond (a measure of how it responds to stress/strain, NOT the breaking point which can be much lower) is about 1x10^12 Pascal. That pressure in tensile mode, if the stress/strain remained linear, would stretch a diamond to twice its length. Diamond won't stretch that much, of course, and will break under tensile strain MUCH sooner than that. It's much stronger under compressive strain, but you still can't expect to approach that kind of pressure and have it hold. Since our pressure is only three orders of magnitude smaller than the Young's modulus, that suggests to me that while a diamond shell might hold, it wouldn't actually hold by a lot. Include the problems of any defects plus the added local pressure if you accidentally bumped the thing, and you're really pushing it.
So: diamond might do it, if you can manufacture a thin spherical shell of the stuff. Steel won't come close. Exotic stuff like carbon nanotubes might be easier to manufacture, but isn't likely to have much better strength/weight ration than diamond for compression. So while it might be possible, even with future materials and fantastic manufacturing processes it's probably cutting it close.
