test: characterize wildcard class escape behavior

Add -#0 self-test cases for backslash escapes inside a '*[...]' class.
They pin two quirks of the current decoder: '\X' matches both X and the
backslash itself, and a literal ']' cannot be a class member because the
parser stops at the first ']' (escaped or not). The latter is why the
filter guide's '*[\[\]]' = "the [ or ] character" claim is wrong (#148):
it parses as the class {[,\} plus a trailing literal ']'. These tests
lock the behavior down so a later matcher fix is a deliberate change.

refs #148

tests/01_engine-filter.test

@@ -47,3 +47,25 @@ match '*foo*bar' 'foozbar'
 
 # '?' is the query-string marker, not a single-char wildcard
 nomatch 'a?c' 'abc'
+
+# backslash escapes a metacharacter inside a class so it is matched literally.
+# Quirk: the decoder also adds the backslash itself to the set, so '\X' matches
+# both X and '\'. These assertions pin that behavior.
+match '*[\*]' '*'
+match '*[\*]' "\\"
+nomatch '*[\*]' 'a'
+match '*[\\]' "\\"
+nomatch '*[\\]' 'a'
+match '*[\[]' '['
+match '*[\[]' "\\"
+nomatch '*[\[]' 'a'
+
+# A literal ']' cannot be a class member: the class parser stops at the first
+# ']', escaped or not. So '*[\[\]]' does NOT mean "the [ or ] character" as the
+# filter guide claims (GitHub #148); it parses as the class {'[','\'} followed
+# by a trailing literal ']'. These assertions document the current (buggy)
+# behavior so any future matcher fix is a deliberate, visible change.
+nomatch '*[\[\]]' '['   # not matched, despite the docs
+match '*[\[\]]' ']'     # only via the empty class-match + trailing ']'
+match '*[\[\]]' '[]'    # one of {'[','\'} then the trailing ']'
+nomatch '*[\[\]]' '[]x'