feat(ml): stricter auto-apply defaults to cut misfires (milestone 139)

head_auto_apply_precision 0.97→0.98, head_auto_apply_min_positives 30→50,
ccip_auto_apply_threshold 0.92→0.95 (operator-asked). Model defaults change for
fresh installs; migration 0081 bumps the existing singleton row IFF still at the
old default (won't clobber a deliberate operator change). ml_admin bounds already
permit these. Fixed a stale comment in the auto-apply test.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM

alembic/versions/0081_stricter_auto_apply_defaults.py

@@ -0,0 +1,50 @@
+"""stricter auto-apply defaults (milestone 139) — cut graduate/auto-apply misfires
+
+head_auto_apply_precision 0.97→0.98, head_auto_apply_min_positives 30→50,
+ccip_auto_apply_threshold 0.92→0.95 (operator-asked 2026-07-06). The model
+defaults change for fresh installs; here we bump the existing singleton row IFF
+it is still at the previous default, so a deliberate operator change is NOT
+clobbered.
+
+Revision ID: 0081
+Revises: 0080
+Create Date: 2026-07-06
+"""
+from typing import Sequence, Union
+
+from alembic import op
+
+revision: str = "0081"
+down_revision: Union[str, None] = "0080"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    op.execute(
+        "UPDATE ml_settings SET head_auto_apply_precision = 0.98 "
+        "WHERE head_auto_apply_precision = 0.97"
+    )
+    op.execute(
+        "UPDATE ml_settings SET head_auto_apply_min_positives = 50 "
+        "WHERE head_auto_apply_min_positives = 30"
+    )
+    op.execute(
+        "UPDATE ml_settings SET ccip_auto_apply_threshold = 0.95 "
+        "WHERE ccip_auto_apply_threshold = 0.92"
+    )
+
+
+def downgrade() -> None:
+    op.execute(
+        "UPDATE ml_settings SET head_auto_apply_precision = 0.97 "
+        "WHERE head_auto_apply_precision = 0.98"
+    )
+    op.execute(
+        "UPDATE ml_settings SET head_auto_apply_min_positives = 30 "
+        "WHERE head_auto_apply_min_positives = 50"
+    )
+    op.execute(
+        "UPDATE ml_settings SET ccip_auto_apply_threshold = 0.92 "
+        "WHERE ccip_auto_apply_threshold = 0.95"
+    )

backend/app/models/ml_settings.py

@@ -51,7 +51,10 @@ class MLSettings(Base):
         Integer, nullable=False, default=8
     )
     head_auto_apply_precision: Mapped[float] = mapped_column(
-        Float, nullable=False, default=0.97
+        # Stricter graduation bar (was 0.97) to cut auto-apply misfires
+        # (operator-asked 2026-07-06): a higher precision target → fewer heads
+        # graduate and those that do get a higher per-head auto_apply_threshold.
+        Float, nullable=False, default=0.98
     )
     # Earned auto-apply (#114). A graduated head fires (tags images without a
     # human) when this master switch is on AND the head has at least
@@ -63,7 +66,9 @@ class MLSettings(Base):
         Boolean, nullable=False, default=True
     )
     head_auto_apply_min_positives: Mapped[int] = mapped_column(
-        Integer, nullable=False, default=30
+        # Support floor raised 30→50 (operator-asked 2026-07-06): a head needs
+        # more human labels before it may fire without a human.
+        Integer, nullable=False, default=50
     )
     # CCIP character-match cosine cut (#114). 0.85 default — the v1 flat 0.75
     # over-fired (high-reference characters matched a scatter of images); 0.85
@@ -78,7 +83,9 @@ class MLSettings(Base):
         Boolean, nullable=False, default=True
     )
     ccip_auto_apply_threshold: Mapped[float] = mapped_column(
-        Float, nullable=False, default=0.92
+        # Raised 0.92→0.95 (operator-asked 2026-07-06) so only very confident
+        # character matches auto-tag.
+        Float, nullable=False, default=0.95
     )
     # Default = SigLIP 2 (so400m, 512px) for new installs (migration 0069);
     # existing libraries keep their stored value until the operator re-embeds.

tests/test_head_auto_apply.py

@@ -88,7 +88,7 @@ def test_sweep_dry_run_counts_but_writes_nothing(db_sync):
 
 
 def test_sweep_skips_under_supported_head(db_sync):
-    # n_pos below head_auto_apply_min_positives (default 30) → a precise-looking
+    # n_pos below head_auto_apply_min_positives (default 50) → a precise-looking
     # but under-supported head never fires.
     img = _img(db_sync, "c" * 64, _emb(0))
     tag = Tag(name="weaktag", kind=TagKind.general)