Compute ScalarFunction properties including return_type and nullable on creation #13825

#[derive(Clone, PartialEq, Eq, PartialOrd, Hash, Debug)]
pub struct ScalarFunction {
    /// The function
    pub func: Arc<crate::ScalarUDF>,
    /// List of expressions to feed to the functions as arguments
    pub args: Vec<Expr>,
    
    // New fields for ScalarFunction
    pub return_type: Vec<DataType>,
    pub nullable: bool,
}

ScalarFunction::new(udf: Arc<crate::ScalarUDF>, args: Vec<Expr>) -> Result<()> {
  1. Check udf's signature against args including length check, type check, implicit coercion
  2. Call fun.return_type() to determine the return type (or even return Field that includes the nullability)
  
  return error if anything goes wrong
}

/// Create a physical expression for the UDF.
pub fn create_physical_expr(
    fun: &ScalarUDF,
    input_phy_exprs: &[Arc<dyn PhysicalExpr>],
    input_schema: &Schema,
    args: &[Expr],
    input_dfschema: &DFSchema,
) -> Result<Arc<dyn PhysicalExpr>> {
    let input_expr_types = input_phy_exprs
        .iter()
        .map(|e| e.data_type(input_schema))
        .collect::<Result<Vec<_>>>()?;

    // verify that input data types is consistent with function's `TypeSignature`
    data_types_with_scalar_udf(&input_expr_types, fun)?;

    // Since we have arg_types, we dont need args and schema.
    let return_type =
        fun.return_type_from_exprs(args, input_dfschema, &input_expr_types)?;

    Ok(Arc::new(
        ScalarFunctionExpr::new(
            fun.name(),
            Arc::new(fun.clone()),
            input_phy_exprs.to_vec(),
            return_type,
        )
        .with_nullable(fun.is_nullable(args, input_dfschema)),
    ))
}