Test with different queue sizes

[j-hui]

Having a size of 8192 means that we necessarily have at least 8192 repeated entries in the log before we encountere a contqueue full error. Can we consider dropping the queue size down to 128 or something so that the event trace is somewhat more easily auditable?

[Rewbert]

We can, I increased it to be able to run some specific tests quite a while back. I don't know what the optimal size is though, if there is any.

[j-hui]

If there are reasons for setting the size larger, it sounds like this is something that we might even want to vary for the purposes of testing.

[j-hui]

A nice-to-have feature that should be fairly straightforward to support is to run each test multiple times, with different queue sizes.

[Rewbert]

Modified the interpreter on branch #interpretable to be parameterised over the queue sizes. Should solve half of that feature ;-)

[j-hui]

Closed by #42