T2: Pydantic v2 — replace @dataclass with BaseModel

- All domain models now use pydantic.BaseModel
- Removed manual to_dict()/from_dict() — use model_dump(mode='json')/model_validate()
- Added field validators:
  - Account.balance >= 0
  - Transaction.amount != 0
  - RecurringCashflow.amount != 0, frequency in {daily, weekly, monthly, yearly}
  - Liability.interest, payment >= 0
  - ExchangeRate.rate > 0
  - ForecastScenario.multipliers >= 0
- pydantic v2 native UUID handling (auto str in JSON)
- Updated excel_sync.py to use model_validate()
- Updated tests/test_model.py and tests/test_currency.py
- Added pydantic>=2.0 to pyproject.toml

Tests: 63/63 pass.

domain/scenario.py

@@ -1,33 +1,19 @@
-from dataclasses import dataclass, field
 from uuid import UUID, uuid4
 
+from pydantic import BaseModel, Field, field_validator
 
-@dataclass
-class ForecastScenario:
-    id: UUID = field(default_factory=uuid4)
+
+class ForecastScenario(BaseModel):
+    id: UUID = Field(default_factory=uuid4)
     name: str = "baseline"
     income_multiplier: float = 1.0
     expense_multiplier: float = 1.0
     growth_multiplier: float = 1.0
     description: str = ""
 
-    def to_dict(self) -> dict:
-        return {
-            "id": str(self.id),
-            "name": self.name,
-            "income_multiplier": self.income_multiplier,
-            "expense_multiplier": self.expense_multiplier,
-            "growth_multiplier": self.growth_multiplier,
-            "description": self.description,
-        }
-
+    @field_validator("income_multiplier", "expense_multiplier", "growth_multiplier")
     @classmethod
-    def from_dict(cls, data: dict) -> "ForecastScenario":
-        return cls(
-            id=UUID(data["id"]),
-            name=data["name"],
-            income_multiplier=data.get("income_multiplier", 1.0),
-            expense_multiplier=data.get("expense_multiplier", 1.0),
-            growth_multiplier=data.get("growth_multiplier", 1.0),
-            description=data.get("description", ""),
-        )
+    def _non_negative(cls, v: float) -> float:
+        if v < 0:
+            raise ValueError("multiplier must be non-negative")
+        return v