Fix inconsistent tiebreak scores when --nth is used

Make sure to consistently calculate tiebreak scores based on the
original line.

This change may not be preferable if you filter aligned tabular input on
a subset of columns using --nth. However, if we calculate length
tiebreak only on the matched components instead of the entire line, the
result can be very confusing when multiple --nth components are
specified, so let's keep it simple and consistent.

Close #926

src/item.go

@@ -7,6 +7,7 @@ import (
 // Item represents each input line
 type Item struct {
 	index       int32
+	trimLength  int32
 	text        util.Chars
 	origText    *[]byte
 	colors      *[]ansiOffset
@@ -18,6 +19,14 @@ func (item *Item) Index() int32 {
 	return item.index
 }
 
+func (item *Item) TrimLength() int32 {
+	if item.trimLength >= 0 {
+		return item.trimLength
+	}
+	item.trimLength = int32(item.text.TrimLength())
+	return item.trimLength
+}
+
 // Colors returns ansiOffsets of the Item
 func (item *Item) Colors() []ansiOffset {
 	if item.colors == nil {