tests: second chimera sequence on TRG

Again, this sequence should not be segmented on IGL, but reported as AMBIGUOUS on TRG and not TOO_FEW_J.

algo/tests/chimera.should_get

@@ -4,4 +4,5 @@ $ Do not segment on IGL by chance
 1:IGL .* 0   
 
 $ Report the sequence as AMBIGUOUS
-1:UNSEG ambiguous .* 1
+1:UNSEG ambiguous .* 2
+

data/chimera-trg.fa

->leu-trg-chimera
+>leu-trg-chimera-1
 ATTCTGGGGTCTATTACTGTGCCACCTGGGACAGGCCCCGGC
 AAGATATTTGCTGAAGGGACTAAGCTCATAGTAACTTCGCCT
 GGTAAGGAAGGCCCCACAGCGTCTTCTGTACTATGACGTCTC
 CACCGCAAGGGATGTGTTGGAATCAGGACTCAGTCCAGGAAA
 GTATTATACTCATACACCCAGGAGGTGGAGCTGGATATTGAG
 ACTGCAAAATCTAATTGAAAATGATTCTGGG
+
+>leu-trg-chimera-2
+TGGGACAGGCCCCGGCAAGATATTTGCTGAAGGGACTAAGCT
+CATAGAACTTCGCCTGGTAAGGAAGGCCCCACAGCATCTTCT
+GTACTATGACGTCTCCACCGCAAGGGATGTGTTGGAATCAGG
+ATCTCAGGT