Extract Krea POV support helpers

docs/prompt-architecture-improvement-plan.md

@@ -149,11 +149,13 @@ Already isolated:
 - `krea_action_context.py` owns shared action-family predicates, axis context
   text, climax detection, and detail-density normalization used by action and
   POV formatter routes.
+- `krea_pov.py` owns POV labels, POV label filtering, and POV camera/composition
+  support text.
 
 Improve later:
 
 - split semantic blocks into modules:
-  `krea_actions.py`, `krea_pov.py`;
+  `krea_actions.py`, `krea_pov_actions.py`;
 - add route-level smoke fixtures for representative metadata rows;
 - make `_hardcore_action_sentence` dispatch by action family instead of long
   conditional chains.

docs/prompt-pool-routing-map.md

@@ -548,6 +548,7 @@ Key Krea2 ownership:
 - Cast descriptor naturalization: `krea_cast.cast_prose`,
   `krea_cast.natural_label_text`.
 - Action context and family predicates: `krea_action_context.py`.
+- POV labels, filtering, and camera/composition support: `krea_pov.py`.
 - Hardcore action sentence: `_hardcore_action_sentence`.
 - POV hardcore sentence: `_pov_hardcore_pose_sentence`, `_pov_action_phrase`.
 - Clothing state cleanup: `krea_clothing.natural_clothing_state`.
@@ -744,9 +745,11 @@ Use these traces to narrow a problem in one pass.
 2. Confirm Krea input uses metadata, not plain prompt fallback.
 3. Inspect `source_role_graph`, `item`, `source_composition`, and
    `item_axis_values`.
-4. Edit `_pov_hardcore_pose_sentence` if the first-person body geometry is
+4. Inspect `krea_pov.py` if the label omission, camera phrase, or POV
+   composition cleanup is wrong.
+5. Edit `_pov_hardcore_pose_sentence` if the first-person body geometry is
    wrong.
-5. Edit `sexual_poses.json` if the raw action lacks enough body-position anchor
+6. Edit `sexual_poses.json` if the raw action lacks enough body-position anchor
    for any formatter to infer a good POV prompt.
 
 ### Camera disappears or becomes too generic