Add JSON schemas, data validator, and first CE1 content

Introduces a generic JSON-Schema-subset validator in GDScript
(scripts/data/schema_validator.gd) used both by `make validate-data`
and, later, by ExerciseEngine at load time, plus semantic_checks.gd
for cross-field constraints a schema can't express (answer index
bounds, at least one faulty token in sentence correction, word count
matching between a dictation's target_text and its annotations).

Schemas cover rule, exercise (dispatched per type: multiple_choice,
sentence_correction), dictation, curriculum, and save (save.schema.json
is not yet wired into validate-data; saves live in user://, handled
at the Profiles/Saves step).

First draft content for the CE1 "accord_adjectif" rule (one rule, two
multiple-choice exercises, one sentence correction, one dictation,
one curriculum entry) exercises the full pipeline end to end and
seeds docs/review/content-review.md, regenerated by the new
tools/generate_content_review.py (`make content-review`).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

scripts/data/validate_data_cli.gd

@@ -1,46 +1,68 @@
 extends SceneTree
 ## Outil headless invoque par `make validate-data`.
-## A ce stade, verifie seulement que les fichiers data/**/*.json sont du JSON bien forme.
-## La validation contre les schemas (rule, exercise, dictation, curriculum, save) sera
-## ajoutee a l'etape "Donnees" (voir Ordre de construction, docs/specs/decisions.md point 2).
+## Valide chaque fichier JSON de data/ contre son schema (data/schemas/), puis
+## applique les contraintes croisees que le schema ne peut pas exprimer.
 
 
 func _init() -> void:
-    var json_paths: Array[String] = _find_json_files("res://data")
     var failures: Array[String] = []
-    for path in json_paths:
-        var file := FileAccess.open(path, FileAccess.READ)
-        if file == null:
-            failures.append("%s : lecture impossible" % path)
-            continue
-        var text := file.get_as_text()
-        if text.strip_edges().is_empty():
-            continue
-        if JSON.parse_string(text) == null:
-            failures.append("%s : JSON invalide" % path)
+    var checked := 0
+
+    checked += _validate_simple_dir("res://data/rules", "rule", failures)
+    checked += _validate_simple_dir("res://data/curriculum", "curriculum", failures)
+    checked += _validate_dictations("res://data/dictations", failures)
+    checked += _validate_exercises("res://data/exercises", failures)
 
     if failures.is_empty():
-        print("validate-data : %d fichier(s) JSON valide(s)" % json_paths.size())
+        print("validate-data : %d fichier(s) JSON valide(s)" % checked)
         quit(0)
     else:
         for failure in failures:
             printerr(failure)
+        printerr("validate-data : %d erreur(s)" % failures.size())
         quit(1)
 
 
-func _find_json_files(root_path: String) -> Array[String]:
-    var result: Array[String] = []
-    var dir := DirAccess.open(root_path)
-    if dir == null:
-        return result
-    dir.list_dir_begin()
-    var entry := dir.get_next()
-    while entry != "":
-        if entry != "." and entry != "..":
-            var full_path := root_path.path_join(entry)
-            if dir.current_is_dir():
-                result.append_array(_find_json_files(full_path))
-            elif entry.ends_with(".json"):
-                result.append(full_path)
-        entry = dir.get_next()
-    return result
+func _validate_simple_dir(dir_path: String, schema_name: String, failures: Array[String]) -> int:
+    var count := 0
+    for path in DataLoader.find_json_files(dir_path):
+        count += 1
+        failures.append_array(DataLoader.validate_file(path, schema_name))
+    return count
+
+
+func _validate_dictations(dir_path: String, failures: Array[String]) -> int:
+    var count := 0
+    for path in DataLoader.find_json_files(dir_path):
+        count += 1
+        var schema_errors := DataLoader.validate_file(path, "dictation")
+        failures.append_array(schema_errors)
+        if schema_errors.is_empty():
+            var data: Dictionary = DataLoader.read_json(path)
+            failures.append_array(SemanticChecks.check_dictation(data, path))
+    return count
+
+
+func _validate_exercises(dir_path: String, failures: Array[String]) -> int:
+    var count := 0
+    for path in DataLoader.find_json_files(dir_path):
+        count += 1
+        var data: Variant = DataLoader.read_json(path)
+        if data == null or typeof(data) != TYPE_DICTIONARY or not data.has("type"):
+            failures.append("%s : champ 'type' manquant ou JSON invalide" % path)
+            continue
+        var exercise_type: String = data["type"]
+        var schema_name := DataLoader.exercise_schema_name_for_type(exercise_type)
+        if schema_name.is_empty():
+            failures.append("%s : type d'exercice inconnu '%s'" % [path, exercise_type])
+            continue
+        var schema_errors := DataLoader.validate_file(path, schema_name)
+        failures.append_array(schema_errors)
+        if not schema_errors.is_empty():
+            continue
+        match exercise_type:
+            "multiple_choice":
+                failures.append_array(SemanticChecks.check_multiple_choice(data, path))
+            "sentence_correction":
+                failures.append_array(SemanticChecks.check_sentence_correction(data, path))
+    return count