Add the playable scene loop: boot through mission to reward

Wires every autoload up to a real UI: boot -> profile select/create
-> world map (three mission markers, positioned from data, not
hardcoded pixels) -> mission intro (narrative from data/missions/,
a new data type with its own schema) -> rule panel -> exercise host
-> one of the three exercise type scenes -> feedback panel -> reward
screen -> back to the map.

exercise_host is the only piece with real orchestration logic: picks
exercises via ExerciseEngine (now fully implemented, validating at
load like make validate-data does), validates answers through the
pure AnswerValidator/DictationAnalyzer, escalates aid over 3 attempts
via AttemptFlow, feeds DifficultyManager so adaptive difficulty is
actually wired in rather than dead code, and records every attempt
through ProgressManager. Every exercise-type scene (multiple_choice,
sentence_correction, dictation) only emits its answer by signal, per
the architecture invariant.

All child-facing text now comes from data/localization/fr.json via
UIText, not hardcoded in scripts or scenes. Getting that right
required extending SchemaValidator to support "additionalProperties"
as a schema (dynamic keys), which also closed a real gap: save.skills
was previously unvalidated against SkillState's actual shape.

16 new headless tests, including two integration tests that instance
exercise_host.tscn directly and drive it by emitting child-scene
signals, stopping short of the final scene transition (which would
tear down the test runner's own tree). Everything that needs a
screen -- layout at different sizes, touch, virtual keyboard, actual
audio playback (no .ogg files exist yet) -- is listed in the new
docs/testing/manual-checklist.md instead of claimed as verified.

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

scenes/exercises/exercise_host.gd

@@ -0,0 +1,150 @@
+extends Control
+## Hote d'exercice : charge la regle, enchaine les exercices d'une mission,
+## valide chaque reponse via les validateurs purs (AnswerValidator /
+## DictationAnalyzer), applique l'aide graduee (AttemptFlow) et enregistre
+## chaque tentative dans ProgressManager (docs/specs/decisions.md, point 7).
+## Ne charge et ne sauvegarde rien lui-meme sur disque : ProgressManager et
+## SaveManager restent seuls responsables de la persistance.
+
+const MISSION_LENGTH := 3
+
+const TYPE_SCENES := {
+    "multiple_choice": "res://scenes/exercises/multiple_choice.tscn",
+    "sentence_correction": "res://scenes/exercises/sentence_correction.tscn",
+    "dictation": "res://scenes/exercises/dictation.tscn",
+}
+
+var _mission: Dictionary = {}
+var _rule: Dictionary = {}
+var _skill_id: String = ""
+var _current_exercise: Dictionary = {}
+var _current_exercise_scene: Control = null
+var _current_difficulty := 1
+var _attempt_number := 1
+var _completed_count := 0
+var _last_exercise_id := ""
+
+@onready var _exercise_container: Control = %ExerciseContainer
+@onready var _overlay_container: Control = %OverlayContainer
+
+
+## Accesseurs reserves aux tests d'integration (scenes/exercises/
+## exercise_host.tscn n'expose pas de script "ami" en GDScript).
+func get_current_exercise() -> Dictionary:
+    return _current_exercise
+
+
+func get_attempt_number() -> int:
+    return _attempt_number
+
+
+func get_completed_count() -> int:
+    return _completed_count
+
+
+func _ready() -> void:
+    _mission = ExerciseEngine.get_mission(GameManager.pending_mission_id)
+    _rule = ExerciseEngine.get_rule(_mission["rule_id"])
+    _skill_id = _rule.get("skill", "")
+    _current_difficulty = ProgressManager.next_difficulty_for_skill(_skill_id, 1)
+    _show_rule_panel()
+
+
+func _show_rule_panel() -> void:
+    var rule_panel: Control = preload("res://scenes/ui/rule_panel.tscn").instantiate()
+    _overlay_container.add_child(rule_panel)
+    rule_panel.show_rule(_rule)
+    rule_panel.continued.connect(_on_rule_panel_continued.bind(rule_panel))
+
+
+func _on_rule_panel_continued(rule_panel: Control) -> void:
+    rule_panel.queue_free()
+    _load_next_exercise()
+
+
+func _load_next_exercise() -> void:
+    _current_exercise = ExerciseEngine.pick_exercise_for_mission(
+        _mission, _last_exercise_id, _current_difficulty
+    )
+    _attempt_number = 1
+
+    if _current_exercise_scene != null:
+        _current_exercise_scene.queue_free()
+
+    var scene_path: String = TYPE_SCENES[_mission["exercise_type"]]
+    _current_exercise_scene = load(scene_path).instantiate()
+    _exercise_container.add_child(_current_exercise_scene)
+    _current_exercise_scene.load_exercise(_current_exercise)
+    _connect_answer_signal()
+
+
+func _connect_answer_signal() -> void:
+    match _mission["exercise_type"]:
+        "multiple_choice":
+            _current_exercise_scene.answer_submitted.connect(_on_multiple_choice_submitted)
+        "sentence_correction":
+            _current_exercise_scene.answer_submitted.connect(_on_sentence_correction_submitted)
+        "dictation":
+            _current_exercise_scene.answer_submitted.connect(_on_dictation_submitted)
+
+
+func _on_multiple_choice_submitted(selected_index: int) -> void:
+    _handle_result(AnswerValidator.validate_multiple_choice(_current_exercise, selected_index))
+
+
+func _on_sentence_correction_submitted(selected_indices: Array, corrections: Dictionary) -> void:
+    var typed_indices: Array[int] = []
+    for index: int in selected_indices:
+        typed_indices.append(index)
+    _handle_result(
+        AnswerValidator.validate_sentence_correction(
+            _current_exercise, typed_indices, corrections
+        )
+    )
+
+
+func _on_dictation_submitted(submitted_text: String) -> void:
+    _handle_result(DictationAnalyzer.analyze(_current_exercise, submitted_text))
+
+
+func _handle_result(result: CorrectionResult) -> void:
+    var support_level := AttemptFlow.SupportLevel.EXPLANATION
+    var is_terminal := result.is_correct
+
+    if not result.is_correct:
+        support_level = AttemptFlow.support_level_for_attempt(_attempt_number)
+        is_terminal = AttemptFlow.should_end_exercise(_attempt_number)
+
+    ProgressManager.record_attempt(
+        _skill_id, result.is_correct, _attempt_number > 1, _mission["exercise_type"]
+    )
+
+    _show_feedback(result, support_level, is_terminal)
+
+
+func _show_feedback(
+    result: CorrectionResult, support_level: AttemptFlow.SupportLevel, is_terminal: bool
+) -> void:
+    var feedback: Control = preload("res://scenes/ui/feedback_panel.tscn").instantiate()
+    _overlay_container.add_child(feedback)
+    feedback.show_result(result, support_level)
+    feedback.continue_pressed.connect(_on_feedback_continued.bind(feedback, is_terminal))
+
+
+func _on_feedback_continued(feedback: Control, is_terminal: bool) -> void:
+    feedback.queue_free()
+
+    if not is_terminal:
+        _attempt_number += 1
+        return
+
+    _current_difficulty = ProgressManager.next_difficulty_for_skill(
+        _skill_id, _current_difficulty
+    )
+    _last_exercise_id = _current_exercise.get("id", "")
+    _completed_count += 1
+
+    if _completed_count >= MISSION_LENGTH:
+        GameManager.goto(GameManager.SCENE_REWARD)
+    else:
+        _load_next_exercise()