Add tags to all entity types with chip-based input and autocomplete

- Add `tags: List[str]` field to all 13 entity types (devices, output targets,
  CSS sources, picture sources, audio sources, value sources, sync clocks,
  automations, scene presets, capture/audio/PP/pattern templates)
- Update all stores, schemas, and route handlers for tag CRUD
- Add GET /api/v1/tags endpoint aggregating unique tags across all stores
- Create TagInput component with chip display, autocomplete dropdown,
  keyboard navigation, and API-backed suggestions
- Display tag chips on all entity cards (searchable via existing text filter)
- Add tag input to all 14 editor modals with dirty check support
- Add CSS styles and i18n keys (en/ru/zh) for tag UI
- Also includes code review fixes: thread safety, perf, store dedup

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

server/src/wled_controller/api/schemas/scene_presets.py

@@ -20,6 +20,7 @@ class ScenePresetCreate(BaseModel):
     name: str = Field(description="Preset name", min_length=1, max_length=100)
     description: str = Field(default="", max_length=500)
     target_ids: Optional[List[str]] = Field(None, description="Target IDs to capture (all if omitted)")
+    tags: List[str] = Field(default_factory=list, description="User-defined tags")
 
 
 class ScenePresetUpdate(BaseModel):
@@ -29,6 +30,7 @@ class ScenePresetUpdate(BaseModel):
     description: Optional[str] = Field(None, max_length=500)
     order: Optional[int] = None
     target_ids: Optional[List[str]] = Field(None, description="Update target list: keep state for existing, capture fresh for new, drop removed")
+    tags: Optional[List[str]] = None
 
 
 class ScenePresetResponse(BaseModel):
@@ -39,6 +41,7 @@ class ScenePresetResponse(BaseModel):
     description: str
     targets: List[TargetSnapshotSchema]
     order: int
+    tags: List[str] = Field(default_factory=list, description="User-defined tags")
     created_at: datetime
     updated_at: datetime