Discuss approach how to reduce python maintenance

@hgy59 and @mreid-tt Over the last few weeks I've update all python packages (pending publishing) from 3.10 up to 3.14 wich is now working well (e.g. `python314-wheels` now enabled by default). Along side to that I've also updated all common wheels used for the crossenv for building, along with examples as per `python31*-wheels` packages. Having 5x python is too much, like way too much. I would like you opinion on how to proceed to reduce the maintenance. But first the theory (and feel free to add to this): ### python310 (end-of-life 2026-10) - Last functional version for old PPC - `numpy-latest` no longer supported ### python311 (end-of-life 2027-10) - Last functional version for armv5 - Current for armv5 `borgbackup` ### python312 (end-of-life 2028-10) - current default ### python313 (end-of-life 2029-10) - current for `homeassistant` ### python314 (end-of-life 2030-10) - current latest official stable version ## Proposal no. 1 - Migrate everything to `python314` - and consider sticking to it for 24-months? - Keep `python311` for armv5 support (for another year, up to end-of-life) - Remove `python310`, `312`, `313` - Start supporting `python315` - no package migration planned, just keeping up and shaking bugs with spksrc. ## Proposal no. 2 - Migrate everything to `python313` - and consider sticking to it for 12-18-months? - Keep `python311` for armv5 support (for another year, up to end-of-life) - Remove `python310`, `312` ... and `314` (in favor of `315` when available). - Start supporting `python315` - no package migration planned, just keeping up and shaking bugs with spksrc. ## Other proposals? Noting that I am completely opened to ideas, it's just too much work as it is.