style(heads): fix import ordering (ruff I001)

Alphabetize HeadTrainingRun in models/__init__ + maintenance imports (H before
I), and drop the inline comment that split heads.py's import block. Pure import
ordering — no behavior change. (run 1601 lint)

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa

backend/app/models/__init__.py

@@ -8,6 +8,7 @@ from .base import Base
 from .credential import Credential
 from .download_event import DownloadEvent
 from .external_link import ExternalLink
+from .head_training_run import HeadTrainingRun
 from .image_prediction import ImagePrediction
 from .image_provenance import ImageProvenance
 from .image_record import ImageRecord
@@ -26,7 +27,6 @@ from .series_suggestion import SeriesSuggestion
 from .source import Source
 from .subscribestar_failed_media import SubscribeStarFailedMedia
 from .subscribestar_seen_media import SubscribeStarSeenMedia
-from .head_training_run import HeadTrainingRun
 from .tag import Tag, TagKind, image_tag
 from .tag_alias import TagAlias
 from .tag_allowlist import TagAllowlist

backend/app/services/ml/heads.py

@@ -34,8 +34,6 @@ from ...models import (
     TagKind,
 )
 from ...models.tag import image_tag
-# Reuse the eval's proven, identical data loaders + metric math so a production
-# head's quality matches what the eval reported for the same concept.
 from .tag_eval import (
     _auto_apply_point,
     _ids_with_tag,

backend/app/tasks/maintenance.py

@@ -13,11 +13,11 @@ from ..celery_app import celery
 from ..models import (
     BackupRun,
     DownloadEvent,
+    HeadTrainingRun,
     ImageRecord,
     ImportBatch,
     ImportSettings,
     ImportTask,
-    HeadTrainingRun,
     LibraryAuditRun,
     Source,
     TagEvalRun,