[stack-switching] Updated bytecode reading to accept switch handling
An update in the stack switching proposal changed the bytecode format for resume and resume_throw. This PR updates the corresponding parts in Wizard.
Actually after checking it seems that the spectests were already updated (compiled with new version ref interpreter). However since most of the tests that involve resume/suspend features only one handler that has a single entry, which handles tag #0 and jumps to the inner-most block of depth 0. This resulted in a lot of tests being passed even when the new code validator isn't implemented (as a lot of the handlers have sequence \00\00\00).
Should I merge this PR, or is its logic included in the #216 ?
#216 is rebased onto this branch. I think we can work on #216 and ignore this for now. Sorry for the confusion.