Group Additivity Double Counting - O2s-O2s(Cds-O2d) and O2s-OsH

Looks like we may be double counting with O2s-O2s(Cds-O2d) and O2s-OsH groups:

example with O=COO
https://rmg.mit.edu/database/thermo/molecule/1%20O%20u0%20p2%20c0%20%7B2,D%7D%0A2%20C%20u0%20p0%20c0%20%7B1,D%7D%20%7B3,S%7D%20%7B5,S%7D%0A3%20O%20u0%20p2%20c0%20%7B2,S%7D%20%7B4,S%7D%0A4%20O%20u0%20p2%20c0%20%7B3,S%7D%20%7B6,S%7D%0A5%20H%20u0%20p0%20c0%20%7B2,S%7D%0A6%20H%20u0%20p0%20c0%20%7B4,S%7D%0A

example with O=C(C)OO
https://rmg.mit.edu/database/thermo/molecule/1%20O%20u0%20p2%20c0%20%7B2,D%7D%0A2%20C%20u0%20p0%20c0%20%7B1,D%7D%20%7B3,S%7D%20%7B4,S%7D%0A3%20C%20u0%20p0%20c0%20%7B2,S%7D%20%7B6,S%7D%20%7B7,S%7D%20%7B8,S%7D%0A4%20O%20u0%20p2%20c0%20%7B2,S%7D%20%7B5,S%7D%0A5%20O%20u0%20p2%20c0%20%7B4,S%7D%20%7B9,S%7D%0A6%20H%20u0%20p0%20c0%20%7B3,S%7D%0A7%20H%20u0%20p0%20c0%20%7B3,S%7D%0A8%20H%20u0%20p0%20c0%20%7B3,S%7D%0A9%20H%20u0%20p0%20c0%20%7B5,S%7D%0A

both GA H298 estimates are about 20 kcal/mol lower than library H298 and the group values are:
O2s-O2s(Cds-O2d): -23.22 kcal/mol
O2s-OsH: -16.3 kcal/mol

Possible solution is to make new "zero" node to avoid adding O2s-OsH when we add O2s-O2s(Cds-O2d) group:

entry(
    index = 227800,
    label = "O2s-(Os-CdOd)H",
    group = 
"""
1 * O2s u0 {2,S} {3,S}
2   O2s u0 {1,S} {4,S}
3   H   u0 {1,S}
4   CO  u0 {2,S} {5,D}
5   O2d u0 {4,D}
""",
    thermo = ThermoData(
        Tdata = ([300,400,500,600,800,1000,1500],'K'),
        Cpdata = ([0,0,0,0,0,0,0],'cal/(mol*K)'),
        H298 = (0,'kcal/mol'),
        S298 = (0,'cal/(mol*K)'),
    ),
    shortDesc = """set to 0 to avoid double counting with O2s-O2s(Cds-O2d)""",
    longDesc = 
"""

""",
)

Start with the two linked molecule examples and compare their group-additivity H298 estimates with the library values, focusing on O2s-O2s(Cds-O2d) and O2s-OsH. Review the proposed zero node and confirm that the correction removes the double counting without changing unrelated estimates.