Protective Chamber (Design & Planning)

You're right to be VERY wary and careful around any compressed gas. The nature of its compressibility makes it such that you are basically dealing with a bomb with the potential for serious damage and destruction. For this reason it is SOP (Standard Operating Procedure) to fill the item to be tested with a fluid first. As you know, most fluids are almost incompressible, so they don't present anywhere near the safety problems if there is a leak or rupture. A compressed gas (air) or a hydraulic pump can then be used to pressurize it for testing. If something does let go, there is very little expansion pressure and force to be released. I've used this method for years in a number of settings safely, including instances where the item being tested did not withstand the testing and ruptured or leaked. If corrosion due to the exposure of your valve to water using this method is a concern, you can utilize either a (hydraulic) oil or antifreeze & water mixture to fill the item being tested.
That being said, nothing wrong with utilizing a containment vessel as well, better to be safe than sorry. Though in my experience it isn't necessary if utilizing this method. I've had cylinders & vessels rupture at over 3,000 psi with no danger. Just be ready to shut the pressure source down very quickly if it does!
Good luck and err on the side of caution.

When in doubt... Farm it out.

A good place to start may be studying a split rim tire cage in an automotive/truck tire shop. They have been tested, tried and true already. Use this as a base line. Then add allowances to compensate for increased conditions. You could also shoot a fire arm of similar impact at a test piece of steel to judge its protective qualities.