Expose core_pool_size, max_pool_size and max_tasks_queued in nodetool tpstats / system_views.thread_pools

It would be nice to expose this information so it is visible from cql(sh) / nodetool tpstats. I briefly checked the complexity of this and it should be quite easy, just add few columns in both and put the sizes to metrics so we can read that.

Without this, I think the only way to know the sizes is to read it from JMX. I just prefer some more handy and faster way to check the sizes so I can adjust them accordingly when necessary.