Clamp adjusted ndistinct to positive integer in estimate_hash_bucketsize().

This avoids a possible divide-by-zero in the following calculation,
and rounding the number to an integer seems like saner behavior anyway.
Assuming IEEE math, the division would yield +Infinity which would get
replaced by 1.0 at the bottom of the function, so nothing really
interesting would ensue; but avoiding divide-by-zero seems like a
good idea on general principles.

Per report from Piotr Stefaniak.  No back-patch since this seems
mostly cosmetic.

src/backend/utils/adt/selfuncs.c

@@ -3541,7 +3541,10 @@ estimate_hash_bucketsize(PlannerInfo *root, Node *hashkey, double nbuckets)
 	 * selectivity of rel's restriction clauses that mention the target Var.
 	 */
 	if (vardata.rel)
+	{
 		ndistinct *= vardata.rel->rows / vardata.rel->tuples;
+		ndistinct = clamp_row_est(ndistinct);
+	}
 
 	/*
 	 * Initial estimate of bucketsize fraction is 1/nbuckets as long as the