feat: Add client extraction, amount cross-validation, and workflow fixes

OCR improvements:
- Extract client data (name, CUI, address) from B2B receipts
- Cross-validate amounts using payment methods and TVA entries
- OCR-tolerant patterns for "TOTAL LEI" with common OCR errors
- Better BON FISCAL vs CHITANTA detection

Backend workflow fixes:
- Fix SQLAlchemy deleted instance error in resubmit/submit workflow
- Add session.refresh() after deleting accounting entries
- Add unapprove endpoint (APPROVED → PENDING_REVIEW)
- Add direction filter for receipt listing

Frontend improvements:
- Fix Vue v-else-if chain broken by Menu component
- Unified OCR Preview layout with values table
- Receipt list filter by direction (plati/incasari)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

data-entry-app/backend/app/schemas/ocr.py

@@ -42,9 +42,15 @@ class ExtractionData(BaseModel):
     payment_methods: List[PaymentMethod] = Field(default=[], description="Payment methods from receipt (CARD, NUMERAR)")
     suggested_payment_mode: Optional[str] = Field(default=None, description="Auto-suggested payment mode based on OCR (casa/banca)")
 
+    # Client data (for B2B receipts - buyer information)
+    client_name: Optional[str] = Field(default=None, description="Client/customer company name")
+    client_cui: Optional[str] = Field(default=None, description="Client CUI/CIF fiscal code")
+    client_address: Optional[str] = Field(default=None, description="Client address")
+
     confidence_amount: float = Field(default=0.0, ge=0, le=1, description="Amount extraction confidence")
     confidence_date: float = Field(default=0.0, ge=0, le=1, description="Date extraction confidence")
     confidence_vendor: float = Field(default=0.0, ge=0, le=1, description="Vendor extraction confidence")
+    confidence_client: float = Field(default=0.0, ge=0, le=1, description="Client extraction confidence")
     overall_confidence: float = Field(default=0.0, ge=0, le=1, description="Overall confidence score")
     raw_text: str = Field(default="", description="Raw OCR text")
     ocr_engine: str = Field(default="", description="OCR engine used: paddleocr or tesseract")