feat(x): X (Twitter) als Bezugsquelle pro Lage

X-Accounts werden analog zu Telegram als Quelle (source_type=x_account)
konfiguriert und pro Lage ueber include_x zugeschaltet. Der Scraper
(feeds/x_parser.py, twscrape) liest Account-Timelines, optional ueber
einen HTTP-Proxy mit Fallback auf direkten Abruf ueber die Server-IP.

- DB-Migration include_x, Pydantic-Modelle, incidents-Router
- Orchestrator-X-Pipeline plus Haiku-Account-Vorselektion
- sources-Router /x/validate, x_account-Typ in Stats und Frontend
- Lage-Einstellungen: X-Toggle neben international und Telegram
- twscrape als Abhaengigkeit

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/models.py

@@ -57,6 +57,7 @@ class IncidentCreate(BaseModel):
     retention_days: int = Field(default=0, ge=0, le=999)
     international_sources: bool = False
     include_telegram: bool = False
+    include_x: bool = False
     visibility: str = Field(default="public", pattern="^(public|private)$")
 
 
@@ -71,6 +72,7 @@ class IncidentUpdate(BaseModel):
     retention_days: Optional[int] = Field(default=None, ge=0, le=999)
     international_sources: Optional[bool] = None
     include_telegram: Optional[bool] = None
+    include_x: Optional[bool] = None
     visibility: Optional[str] = Field(default=None, pattern="^(public|private)$")
 
 
@@ -102,6 +104,7 @@ class IncidentResponse(BaseModel):
     public_mood_updated_at: Optional[str] = None
     international_sources: bool = True
     include_telegram: bool = False
+    include_x: bool = False
     created_by: int
     created_by_username: str = ""
     created_at: str
@@ -130,6 +133,7 @@ class IncidentListItem(BaseModel):
     visibility: str = "public"
     international_sources: bool = True
     include_telegram: bool = False
+    include_x: bool = False
     created_by: int
     created_by_username: str = ""
     created_at: str