Updates to electron algorithms

This MR implements the following changes to the "electron working point" algorithm config:

• switch the expected likelihood string from e.g. "TightLHElectron" or "MediumDNNElectron" to "TightLH" and "MediumDNN", this being much closer to the current setup for most users (who do not recompute the ID decision, which is what the original format of this string is needed for) and to the information that can be found on Egamma recommendation twikis;
• remove the electron "effSF" (which is really the reconstruction SF), and replace it by properly set up reconstruction, ID and isolation SFs;
• handle the differences in Run 2 vs Run 3 efficiencies;
• put a placeholder for ChargeID SFs (needed when the ECIDS tool is used), as they are currently not supported in R22.

The last missing electron SF is the trigger one, this will depend on what the upcoming R22 electron recommendations look like... For now, we could simply hardcode a reasonable choice of electron triggers? Eventually, it should depend on the triggers selected by the user, and I don't know what the best way to handle that is. Update: actually we should prefer the TrigGlobalEfficiencyCorrectionTool, but there's no algorithm for that yet...