Jaeger

🚀 New

* Drop `poetry` and use `uv` for project management. Update GitHub workflows.

✨ Improved

* [207](https://github.com/sdss/jaeger/pull/207) Update the call to determine target offsets.
* [206](https://github.com/sdss/jaeger/pull/206) Added flag `--sea-anemone` to `jaeger configuration random`.

⚙️ Engineering

* Relax Python requirement to `^3.10,<4`.

✨ Improved

* Add ``rot_ref_angle`` option to the FVC command to set the reference rotator angle for fiducial fitting.

🔧 Fixed

* Set the configuration epoch for `DitheredConfiguration` as the epoch of the parent.
* Use the configuration focal scale in all calls to `icrs_from_positioner_dataframe()`.

🚑 Hotfix

* [205](https://github.com/sdss/jaeger/pull/205) Added a hotfix for an issue caused when the back-illuminated fibres don't turn on during and FVC loop (cause under investigation). When light is not detected in the FVC image the code will retry up to three times, each time turning the LEDs off and then on again.

⚙️ Engineering

* Add back `pyarrow` dependency.

🏷️ Changed

* [204](https://github.com/sdss/jaeger/pull/204) Replace the old `confSummary_test` paths which have not become the default. When a configuration or FVC writes a configuration file, it does so to `$SDSSCORE_DIR` with the new formatting (i.e., using thousand and hundred groupings). Additionally, and at least for now, it also writes the same file to `$SDSSCORE_LEGACY_DIR` with the old format (only hundred groupings).

⚙️ Engineering

* Upgrade `sdsstools` to 1.7.1 with support for Numpy 2.0.
* Upgrade `polars` to 1.0.0.

🏷️ Changed

* Disabled `alert_fluid_temperature`.

✨ Improved

* Bumped `kaiju` to 1.4.0 with speed-up improvements.

🔧 Fixed

* Fix import of `os` in `target/tools.py` which was affecting the loading of cloned configurations and possibly other features.