improve db generation performance

Author: johannes-moegerle

src/rydstate/species/species_object.py

@@ -269,6 +269,11 @@ def get_ionization_energy(self, unit: str | None = "hartree") -> PintFloat | flo
             return ionization_energy.magnitude
         return ionization_energy.to(unit, "spectroscopy").magnitude
 
+    @cached_property
+    def ionization_energy_au(self) -> float:
+        """Ionization energy in atomic units (Hartree)."""
+        return self.get_ionization_energy("hartree")
+
     @overload
     def get_corrected_rydberg_constant(self, unit: None = None) -> PintFloat: ...
 
@@ -373,7 +378,7 @@ def calc_nu(
         if n <= nist_n_max and use_nist_data:  # try to use NIST data
             if (n, l, j_tot, s_tot) in self._nist_energy_levels:
                 energy_au = self._nist_energy_levels[(n, l, j_tot, s_tot)]
-                energy_au -= self.get_ionization_energy("hartree")
+                energy_au -= self.ionization_energy_au  # use the cached ionization energy for better performance
                 return calc_nu_from_energy(self.reduced_mass_au, energy_au)
             logger.debug(
                 "NIST energy levels for (n=%d, l=%d, j_tot=%s, s_tot=%s) not found, using quantum defect theory.",