issue#252: analyze start in each creation index thread

Instead of running analyze in a separated transaction after all indexes
are created, now each table is analyzed in the the same transaction
as indexes are created. This way app got informed about 'HAF is ready'
when both indes is created and analyze finished

src/hive_fork_manager/hived_api_impl.sql

@@ -530,6 +530,8 @@ BEGIN
   END LOOP;
   CLOSE __cursor;
 
+  EXECUTE format( 'ANALYZE %s',  _table_name );
+
   DELETE FROM hive.indexes_constraints
   WHERE table_name = _table_name AND is_foreign_key = FALSE;
   RAISE NOTICE 'Finished restoring any dropped indexes on %', _table_name;

src/sql_serializer/indexes_controler.cpp

@@ -94,10 +94,6 @@ indexes_controler::enable_indexes() {
     *restore_applied_hardforks_idxs
   );
 
-  auto analyze_expression_idxs = start_commit_sql( true, "hive.reanalyze_indexes_with_expressions()", "enable indexes" );
-  analyze_expression_idxs->join();
-
-
   fc::time_point cluster_start_time = fc::time_point::now();
   fc::microseconds restore_indexes_time = cluster_start_time - restore_indexes_start_time;
   ilog( "PROFILE: Restored HAF table indexes: ${t}s", ("t",restore_indexes_time.to_seconds()) );