Change dash/mksh unicode test cases to "N-I".

They don't even attempt to implement utf-8 string decoding.

spec/var-op-other.test.sh

@@ -13,15 +13,15 @@ echo ${#v}
v=$'_\u03bc_'
echo ${#v}
## stdout: 3
## BUG dash stdout: 9
## BUG mksh stdout: 4
## N-I dash stdout: 9
## N-I mksh stdout: 4

### Unicode string length (spec/testdata/utf8-chars.txt)
v=$(cat spec/testdata/utf8-chars.txt)
echo ${#v}
## stdout: 7
## BUG dash stdout: 13
## BUG mksh stdout: 13
## N-I dash stdout: 13
## N-I mksh stdout: 13

### String length with incomplete utf-8
for num_bytes in 0 1 2 3 4 5 6 7 8 9 10 11 12 13; do
@@ -60,7 +60,7 @@ error: Incomplete utf-8
9
7
## END
## BUG dash/mksh STDOUT:
## N-I dash/mksh STDOUT:
0
1
2
@@ -115,7 +115,7 @@ error: Invalid start of utf-8 char
10
8
8
## BUG dash STDOUT:
## N-I dash STDOUT:
7
8
9
@@ -132,7 +132,7 @@ error: Invalid start of utf-8 char
20
20
## END
## BUG mksh STDOUT:
## N-I mksh STDOUT:
1
2
3