Return >= 1 from matching_candidates_in_table if records > 0.0

Having rows >= 1.0 helps ensure that when we calculate total rows of joins
the number of resulting rows will not be less after the join.

Changes in test cases:
- Join order change for some tables with few records
- 'Filtered' is much higher for tables with few rows, as 1 row is a high
  procent of a table with few rows.

mysql-test/main/statistics_json.result

@@ -4643,7 +4643,7 @@ Warnings:
 Note	1003	select `test`.`t1_json`.`a` AS `a` from `test`.`t1_json` where `test`.`t1_json`.`a` < 'b-1a'
 analyze select * from t1_json where a > 'zzzzzzzzz';
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
-1	SIMPLE	t1_json	ALL	NULL	NULL	NULL	NULL	10	10.00	0.00	0.00	Using where
+1	SIMPLE	t1_json	ALL	NULL	NULL	NULL	NULL	10	10.00	10.00	0.00	Using where
 drop table ten;
 UPDATE mysql.column_stats 
 SET histogram='["not-what-you-expect"]' WHERE table_name='t1_json';