Refactor of anchor configuration logic

[sawenzel]

This commit refactors the transfer of anchoring configuration from async-reco scripts into the MC pipeline, addressing O2-5011 by replacing the previous whitelisting mechanism with a blacklisting approach.

In addition, it introduces broader configuration/customization support: users can now inject arbitrary command-line options into any task via a customize.json file, reducing the need to edit workflow scripts.

For example:
{
"ConfigParams": {
"EMCSimParam": {
"mBusyTime": "11.11"
}
},
"Executables": {
"o2-ft0-reco-workflow": {
"filtered": {
"--my-custom-option": "foo"
}
},
"ft0fv0emcctp_digi": {
"filtered": {
"--another-custom-option": "baz"
}
}
}
}

This allows for flexible injection of both configKeyValues and CLI options via o2dpg_sim_workflow.py --overwrite-config customize.json.

[github-actions]

REQUEST FOR PRODUCTION RELEASES:
To request your PR to be included in production software, please add the corresponding labels called "async-" to your PR. Add the labels directly (if you have the permissions) or add a comment of the form (note that labels are separated by a ",")

+async-label <label1>, <label2>, !<label3> ...

This will add <label1> and <label2> and removes <label3>.

The following labels are available
async-2023-pbpb-apass4
async-2023-pp-apass4
async-2024-pp-apass1
async-2022-pp-apass7
async-2024-pp-cpass0
async-2024-PbPb-apass1
async-2024-ppRef-apass1
async-2024-PbPb-apass2
async-2023-PbPb-apass5

[sawenzel]

@shahor02 : Proposal for the blacklisting refactor. We will now pickup things like --trackqc-fraction for the aod-converter from the async workflows and in addition we gain a much better configurability overall.

We still need to do some final adjustments, proper testing, and the treatment for TPC distortions. For now WIP.

[shahor02]

@sawenzel thanks, looks nice! (did not dive into the implementation details)

[shahor02]

@sawenzel suppression of this line leads to disabling corrections in the reconstruction even if distortions were used in the digitization (this happened in the recent test of Mesut LHC25b8a_default_v4, while in LHC25b8a_default_v3 they were OK).
I remember you did something else to detect distortion used in the digi. but I don't remember what and it seems it did not work. Shall we simply restore this line?