Openacc tweaks and cleanup 2 (#1099)

* Making all end parallel directives specify end loop

* Replacing last acc declare in a procedure with acc enter/exit data commands

* Removing pure declarations, turns out they dont really improve performance and openmp isn't allowed within them.

* Adding explicit directives to penta_lu and tridiag_lu, as opposed to the previously used kernels directive. Also splitting up a loop to improve GPU performance.

* Splitting up a couple loops for performance reasons.

* Missed one somehow

src/CLUBB_core/adg1_adg2_3d_luhar_pdf.F90

@@ -1360,7 +1360,7 @@ subroutine backsolve_Luhar_params( Sk_max, Skx,          & ! In
   end subroutine backsolve_Luhar_params
 
   !=============================================================================
-  pure function max_cubic_root( a_coef, b_coef, c_coef, d_coef ) &
+  function max_cubic_root( a_coef, b_coef, c_coef, d_coef ) &
   result( max_root )
 
     ! Description:

src/CLUBB_core/advance_clubb_core_module.F90

@@ -5157,7 +5157,7 @@ subroutine calc_trapezoid_zm( nz, ngrdcol, gr, variable_zm, variable_zt, &
                                  * ( gr%zm(i,k) - gr%zt(i,k) ) * gr%invrs_dzm(i,k)
         end do
       end do 
-      !$acc end parallel
+      !$acc end parallel loop
 
       return
     end subroutine calc_trapezoid_zm
@@ -5514,7 +5514,7 @@ subroutine set_Lscale_max( ngrdcol, l_implemented, host_dx, host_dy, &
     end subroutine set_Lscale_max
 
 !===============================================================================
-  pure subroutine calculate_thlp2_rad &
+  subroutine calculate_thlp2_rad &
                   ( nz, rcm_zm, thlprcp, radht_zm, & ! Intent(in)
                     clubb_params,                  & ! Intent(in)
                     thlp2_forcing )                  ! Intent(inout)

src/CLUBB_core/advance_helper_module.F90

@@ -2023,7 +2023,7 @@ function vertical_avg( total_idx, rho_ds, field, dz )
   end function vertical_avg
 
   !=============================================================================
-  pure function vertical_integral( total_idx, rho_ds, &
+  function vertical_integral( total_idx, rho_ds, &
                                        field, dz )
 
     ! Description: