feat(memory): complete M.2-M.5 memory upgrade phases with tests

- M.2: ForgettingCurve, MemoryDecay, MemoryReinforcement (selective forgetting)
- M.3: DailyDigestGenerator, ReminderScheduler, ProactiveInformer (proactive reminders)
- M.4: MemoryExtractor with LLM-based memory extraction from conversations
- M.5: MemoryRecallInjector with token budget control for prompt injection
- All phases include comprehensive unit tests (109 tests passing)
- Updated checklist.md to mark all tasks complete

backend/app/services/memory/memory_decay.py

@@ -0,0 +1,81 @@
+"""
+MemoryDecay
+
+Handles memory archiving, deprioritization, and restoration.
+"""
+
+from datetime import UTC, datetime
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from app.models.memory import UserMemory
+
+
+class MemoryDecay:
+    """Handle memory archiving and deprioritization decisions."""
+
+    ARCHIVE_THRESHOLD = 0.2
+    DEPRIORITIZE_THRESHOLD = 0.5
+
+    def evaluate(self, memory: "UserMemory") -> dict:
+        """Evaluate memory and return action recommendation.
+
+        Returns:
+            dict with keys: decay_score, should_archive, should_deprioritize, action
+        """
+        from app.services.memory.forgetting_curve import ForgettingCurve
+
+        curve = ForgettingCurve()
+        decay_score = curve.calculate_decay(memory)
+        archive = decay_score < self.ARCHIVE_THRESHOLD
+        deprioritize = decay_score < self.DEPRIORITIZE_THRESHOLD
+
+        if archive:
+            action = "archive"
+        elif deprioritize:
+            action = "deprioritize"
+        else:
+            action = "keep_active"
+
+        return {
+            "decay_score": decay_score,
+            "should_archive": archive,
+            "should_deprioritize": deprioritize,
+            "action": action,
+        }
+
+    def archive_memory(self, memory: "UserMemory") -> "UserMemory":
+        """Archive a memory (set is_archived=True, reset decay_score to low value).
+
+        Archived memories are moved to cold storage and not included in
+        active reminders or context injection.
+        """
+        memory.is_archived = True
+        memory.decay_score = 0.1  # Very low, will be restored on access
+        memory.archive_at = datetime.now(UTC)
+        return memory
+
+    def deprioritize_memory(self, memory: "UserMemory") -> "UserMemory":
+        """Mark a memory as deprioritized (excluded from active reminders).
+
+        Unlike archival, the memory is still accessible and included in
+        context injection if relevant.
+        """
+        # Just update decay_score, the importance_level already encodes priority
+        from app.services.memory.forgetting_curve import ForgettingCurve
+
+        curve = ForgettingCurve()
+        memory.decay_score = curve.calculate_decay(memory)
+        return memory
+
+    def restore_from_archive(self, memory: "UserMemory") -> "UserMemory":
+        """Restore a memory from archive.
+
+        Resets is_archived=False and decay_score=0.8 (strong retention).
+        The memory is moved back to hot storage.
+        """
+        memory.is_archived = False
+        memory.decay_score = 0.8  # Strong retention after restore
+        memory.last_accessed_at = datetime.now(UTC)
+        memory.archive_at = None
+        return memory