Mixing scheme fails to mix entries from the same material id even when their structures match

[PutuB]

Python version

Python 3.11.11

Pymatgen version

2025.1.23

Operating system version

No response

Current behavior

Hello, I recently tried using the MaterialsProjectDFTMixingScheme() to construct a GGA/GGA+U/R2SCAN phase diagram for Fe-O by following the code on the materials project documentation. However, I noticed that the constructed phase diagram used the mp-13-GGA entry even though mp-13-r2SCAN exists. It seems that the mixing function did not attempt to combine entries for mp-13 because its GGA and R2SCAN calculations employ different numbers of atoms in the cell:

Full Formula (Fe1)
Reduced Formula: Fe
abc   :   2.459206   2.459352   2.459570
angles: 109.469122 109.467063 109.459583
pbc   :       True       True       True
Sites (1)
  #  SP      a    b    c
---  ----  ---  ---  ---
  0  Fe      0    0    0 

Full Formula (Fe2)
Reduced Formula: Fe
abc   :   2.477813   2.477809   4.054295
angles:  90.000038  90.000320 109.469837
pbc   :       True       True       True
Sites (2)
  #  SP      a    b    c
---  ----  ---  ---  ---
  0  Fe    0.5  0.5  0.5
  1  Fe    1    1    0 

Expected Behavior

The ground state at composition Fe (mp-13) has both GGA/GGA+U and R2SCAN entries, so I expected the R2SCAN entry to replace the GGA one. I have also checked that their structures matched using StructureMatcher().fit().

Minimal example

from mp_api.client import MPRester
from pymatgen.analysis.phase_diagram import PhaseDiagram, PDPlotter
from pymatgen.entries.mixing_scheme import MaterialsProjectDFTMixingScheme

with MPRester(my_key) as mpr:

    # Obtain GGA, GGA+U, and r2SCAN ComputedStructureEntry objects
    entries = mpr.get_entries_in_chemsys(elements=["Fe", "O"], 
                                         additional_criteria={"thermo_types": ["GGA_GGA+U", "R2SCAN"]}) 
    
    # Apply corrections locally with the mixing scheme
    scheme = MaterialsProjectDFTMixingScheme()
    corrected_entries = scheme.process_entries(entries, verbose=True)
    
# Construct phase diagram
pd = PhaseDiagram(corrected_entries)
    
# Plot phase diagram
PDPlotter(pd).get_plot()

Relevant files to reproduce this bug

No response