There was about 10 Nm less with stock VVT map. Engine is fully stock. Rev limiter set to 8000 RPM. Car made 0-100 in 5.12 secs on this setup.

The problem is really not UDS - everything is completely standard there, and not locked behind any security.
Just open ISO14229 and have at it.
The problem is the measuring data, like always.

Anyway it's implemented in my tool...

some solid numbers  there

Digging lately in Volvo ESKONF. Looks like all sw. versions use the same pattern and 7 bytes length (50WRHJ also).
To disable rear O2 heater diagnosis (it will still feed current to heater but disable triggering trouble code) just change 4rd bitpair in second byte to 11b.
1st bit pair in 6th byte is ignition coil no. 3. Do not change that
Last 7th byte may differ the most between software versions (there is f.e. EVAP or ECU fan diagnosis - for cars that have one)

Bits:    7,6         5,4         3,2         1,0
------------------------------------------------     
B0       EV3        EV4          ?           EV5
B1       LS2B1H     LDR          EV2         EV1
B2       A/C        FP           LS2B2H      CP
B3       ?          ?            FTP         FC
B4       ZUE5       ZUE4         ZUE2        ZUE1
B5       ?          ?            ?           ZUE3
B6       ?          ?            ?           MIL


Where:
EV - injectors
ZUE - ignition coils
A/C - air con relay
FP - fuel pump relay
LS2B1H - rear lambda bank1 heater
LS2B2H - rear lambda bank2 heater
CP - canister purge
FC - cooling fan
FTP - fuel tank pressure sensor

I checked 20KPSC, 50QHHJ and 50WRHJ - all have the same pattern except last B6.
Nevertheless all that will ever be needed is eventually disabling rear lambda heater so all other ESKONF items doesn't matter but it was fun to reverse engineer it

Disabling rear lambda heater can also be done using diagnostic trouble codes table.
ME7 algorithm first checks ESKONF and if diagnostics is allowed then it eventually jumps to DTC triggering routine.
DTC triggering routine checks trouble codes table (starting from CLAAAA) if particular DTC code class is "active" (not zero).

Also DTC triggering can be fired from other places in the code so it probably is better to disable DTC triggering.
For example CDHSH codeword specify if rear O2 heating diagnosis should be done at all.
But ECU will stil be able to set DTC by power stage diagnosis (ESKONF)

Summing up, just disable diagnostics trouble table class item

I agree ECU can still diagnose things and react "silently" if error detected.
That's why usually neither disabling ESKONF or disabling error class is necessary and shouldn't be done.
I think rear O2 heating can be exception from this rule. I see no point in leaving DTC for this particular error class especially if you remove rear lambda completely

Do you know description of error class? I tried to restore boost after overboost error. Is it relates to error type? Or is it must be done with healing path for this error?

got an interesting read today.

later 2004 2.5t with fps / pem running the 50QKHJ  never seen that on a non 04R