js-on-sx: built-in .length returns spec-defined values

String.fromCharCode.length, Math.max.length, Array.from.length
were returning 0 because their SX lambdas use &rest args with no
required params — but spec assigns each a specific length.
Added js-builtin-fn-length mapping JS name to spec length (12
entries). js-fn-length consults the table first and falls back to
counting real params.
built-ins/String: 79/99 → 80/99, built-ins/Array: 20/45 → 21/45.
conformance.sh: 148/148.

lib/js/test262-scoreboard.md

@@ -1,24 +1,28 @@
 # test262 scoreboard
 
 Pinned commit: `d5e73fc8d2c663554fb72e2380a8c2bc1a318a33`
-Wall time: 57.4s
+Wall time: 123.5s
 
-**Total:** 44/50 runnable passed (88.0%). Raw: pass=44 fail=3 skip=0 timeout=3 total=50.
+**Total:** 21/45 runnable passed (46.7%). Raw: pass=21 fail=17 skip=5 timeout=7 total=50.
 
 ## Top failure modes
 
-- **3x** Timeout
-- **3x** Test262Error (assertion failed)
+- **14x** Test262Error (assertion failed)
+- **7x** Timeout
+- **2x** TypeError: not a function
+- **1x** Unhandled: Not callable: {:2 43 :1 42 :0 41 :length 3}\
 
 ## Categories (worst pass-rate first, min 10 runnable)
 
 | Category | Pass | Fail | Skip | Timeout | Total | Pass % |
 |---|---:|---:|---:|---:|---:|---:|
-| built-ins/Number | 44 | 3 | 0 | 3 | 50 | 88.0% |
+| built-ins/Array | 21 | 17 | 5 | 7 | 50 | 46.7% |
 
 ## Per-category top failures (min 10 runnable, worst first)
 
-### built-ins/Number  (44/50 — 88.0%)
+### built-ins/Array  (21/45 — 46.7%)
 
-- **3x** Timeout
-- **3x** Test262Error (assertion failed)
+- **14x** Test262Error (assertion failed)
+- **7x** Timeout
+- **2x** TypeError: not a function
+- **1x** Unhandled: Not callable: {:2 43 :1 42 :0 41 :length 3}\