wqzhou

I double-check it, and find the bug is not here.

I make some calculations with different `mixing_beta=0.4, 0.3, 0.2`, and find only `mixing_beta=0.4` with `STRU1` converges.  see more in [tests](https://app.bohrium.dp.tech/abacustest/?request=GET%3A%2Fapplications%2Fabacustest%2Fjobs%2Fjob-ni-abacustest-v0.3.113-07bfd6).

I also try different MPI (OMP_NUM_THREADS=1 mpirun -np 16 & OMP_NUM_THREADS=2 mpirun -np 8) with `STRU1` and `STRU2`, and the result is different. ABACUS shows obvious numerical instability in this...

@dyzheng @jinzx10 any idea?

After PR #4842, the `drho` now is: 

@denghuilu could you have a look?

Since this issue cannot be reproduced now, we close it now. It can be reopened once the bug occurs again.

This issue is from the machine issue, not related with ABACUS.

@AsTonyshment Actually, `LOC` class has been replaced by `DensityMatrix` class, and only use for `out_dm` and EXX calculations now. I think this class will be deleted soon. Why do you...

@QuantumMisaka do you know how make it?