Add profile and save persistence, wired into the three autoloads

scripts/profiles/: SaveIO handles real file I/O in user://saves (one
file per profile_id, atomic write via temp-file-then-rename, backup
copy before each overwrite, corrupted-main falls back to backup,
schema-invalid content is rejected like corruption). SaveMigration
and SaveFactory are pure; SaveMigration refuses a schema_version it
doesn't recognize (older or newer) rather than guessing. ProgressUpdates
is pure and wires ReviewScheduler/DifficultyManager onto a save's
skills dict, including per-session overdue-review marking.

The three autoloads (SaveManager, ProfileManager, ProgressManager)
stay thin: in-memory current-profile state and orchestration only,
no logic of their own, so the actual behavior lives in the tested
pure/file-IO modules above rather than in untestable glue.

26 new headless tests, including the failure modes called out in the
launch prompt: a truncated .tmp file never corrupts the readable
save, a corrupted main file falls back to its backup, two profiles
never mix (down through the real autoload layer, not just SaveIO),
and an unknown future schema_version is rejected outright.

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

autoload/progress_manager.gd

@@ -1,3 +1,58 @@
 extends Node
-## Competences, missions debloquees et planification des revisions par profil.
-## Logique detaillee a partir de l'etape "Logique pure" (voir docs/specs/decisions.md, point 5).
+## Competences, missions debloquees et planification des revisions du profil
+## selectionne. Delegue les calculs a ProgressUpdates (scripts/profiles/), qui
+## s'appuie sur ReviewScheduler et DifficultyManager (scripts/pedagogy/).
+## L'ecriture sur disque reste groupee : save_now() est appele en fin de
+## mission, pas a chaque tentative (docs/specs/decisions.md, point 7).
+
+var _current_save: Dictionary = {}
+var _current_profile_id: String = ""
+
+
+## Charge la sauvegarde existante d'un profil et demarre une nouvelle seance
+## (voir ProgressUpdates.start_session). Faux si le profil n'a pas de
+## sauvegarde lisible.
+func load_profile(profile_id: String) -> bool:
+    var data: Variant = SaveManager.load(profile_id)
+    if data == null:
+        return false
+    _current_profile_id = profile_id
+    _current_save = ProgressUpdates.start_session(data)
+    return true
+
+
+func record_attempt(
+    skill_id: String, success: bool, hint_used: bool, exercise_type: String
+) -> void:
+    var attempt := {
+        "skill": skill_id,
+        "success": success,
+        "hint_used": hint_used,
+        "exercise_type": exercise_type,
+    }
+    _current_save = ProgressUpdates.record_attempt(_current_save, attempt)
+
+
+func unlock_region(region_id: String) -> void:
+    _current_save = ProgressUpdates.unlock_region(_current_save, region_id)
+
+
+func next_difficulty_for_skill(skill_id: String, current_difficulty: int) -> int:
+    return ProgressUpdates.next_difficulty_for_skill(_current_save, skill_id, current_difficulty)
+
+
+func get_skill_state(skill_id: String) -> Dictionary:
+    var skills: Dictionary = _current_save.get("skills", {})
+    return skills.get(skill_id, ReviewScheduler.initial_state())
+
+
+func current_save() -> Dictionary:
+    return _current_save
+
+
+## A appeler en fin de mission (pas apres chaque tentative) pour persister la
+## progression du profil courant.
+func save_now() -> bool:
+    if _current_profile_id.is_empty():
+        return false
+    return SaveManager.save(_current_profile_id, _current_save)