#28 Configurable number of data nodes for regression test cases

[Abbas Butt]

This is required so that if an EXPLAIN for a remote query reports all the data nodes in any regression test case, the expected output can be changed according to the cluster configuration.
See e.g. commented test cases in create_index test case

EXPLAIN (COSTS OFF)
SELECT * FROM fast_emp4000
WHERE home_base @ '(200,200),(2000,1000)'::box
ORDER BY (home_base[0])[0];

This also applies to the number of coordinators.