solve_cg: use linop.make instead of MF constructor (AMReX-Codes#3627)

## Summary

This PR replaces the explicit use of `MF` constructors in
`MLCGSolverT<MF>::solve_cg` with calls to the make method of the linear
operator associated with the `MLCGSolverT` object.

## Additional background

This is a similar to the PR on `solve_bicgstab`.

Src/LinearSolvers/MLMG/AMReX_MLCGSolver.H

@@ -261,17 +261,13 @@ MLCGSolverT<MF>::solve_cg (MF& sol, const MF& rhs, RT eps_rel, RT eps_abs)
 
  const int ncomp = sol.nComp();
 
- const BoxArray& ba = sol.boxArray();
- const DistributionMapping& dm = sol.DistributionMap();
- const auto& factory = sol.Factory();
-
- MF p(ba, dm, ncomp, sol.nGrowVect(), MFInfo(), factory);
+ MF p = Lp.make(amrlev, mglev, sol.nGrowVect());
  p.setVal(RT(0.0));
 
- MF sorig(ba, dm, ncomp, nghost, MFInfo(), factory);
- MF r (ba, dm, ncomp, nghost, MFInfo(), factory);
- MF z (ba, dm, ncomp, nghost, MFInfo(), factory);
- MF q (ba, dm, ncomp, nghost, MFInfo(), factory);
+ MF sorig = Lp.make(amrlev, mglev, nghost);
+ MF r  = Lp.make(amrlev, mglev, nghost);
+ MF z  = Lp.make(amrlev, mglev, nghost);
+ MF q  = Lp.make(amrlev, mglev, nghost);
 
  sorig.LocalCopy(sol,0,0,ncomp,nghost);