Two new tests are skipped because they're buggy. Making some newer
random state tests work required implementing the 'compressed return
value encoding'.
* Removed some strange code that didn't apply state reverting properly
* Refactored code setting from vm & state transition to the executioner
* Updated tests
Log filtering is now using a MIPmap like approach where addresses of
logs are added to a mapped bloom bin. The current levels for the MIP are
in ranges of 1.000.000, 500.000, 100.000, 50.000, 1.000. Logs are
therefor filtered in batches of 1.000.