change source interval in the export package

[iluise]

Description

implement a more solid way to retrieve the forecast steps that takes into account also the substeps.

init_time = valid_time - source_end

Where source_end is the last valid_time of the source window.

example:

uv run export --run-id b9i7napw --output-dir ./results/b9i7napw/ --format quaver --stream ERA5 --type prediction --samples 0 1 --fsteps 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32  --quaver-template-folder ./grib_templates/ --quaver-template-grid-type o96 --expver j3xx --n-processes 48 --channels 2t msl 10u 10v t_500 t_850 z_500 z_850 q_500 q_850 v_500 v_850 u_500 u_850

Dimensions:        (time: 2, step: 60, isobaricInhPa: 2, values: 40320)
Coordinates:
  * time           (time) datetime64[ns] 16B 2023-10-01T05:00:00 2023-10-01T1...
  * step           (step) timedelta64[ns] 480B 07:00:00 ... 5 days 00:00:00
    valid_time     (time, step) datetime64[ns] 960B ...
  * isobaricInhPa  (isobaricInhPa) float64 16B 850.0 500.0
    latitude       (values) float64 323kB ...
    longitude      (values) float64 323kB ...
Dimensions without coordinates: values
Data variables:
    q              (time, step, isobaricInhPa, values) float32 39MB ...
    u              (time, step, isobaricInhPa, values) float32 39MB ...
    v              (time, step, isobaricInhPa, values) float32 39MB ...
    z              (time, step, isobaricInhPa, values) float32 39MB ...
    t              (time, step, isobaricInhPa, values) float32 39MB ...
Attributes:
    GRIB_edition:            2
    GRIB_centre:             ecmf
    GRIB_centreDescription:  European Centre for Medium-Range Weather Forecasts
    GRIB_subCentre:          0
    Conventions:             CF-1.7
    institution:             European Centre for Medium-Range Weather Forecasts
    history:                 2026-04-24T10:40 GRIB to CDM+CF via cfgrib-0.9.1...

steps:

> ds.step.astype("timedelta64[ns]")/3600000000000
<xarray.DataArray 'step' (step: 60)> Size: 480B
array([  7,   8,   9,  10,  11,  12,  19,  20,  21,  22,  23,  24,  31,
        32,  33,  34,  35,  36,  43,  44,  45,  46,  47,  48,  55,  56,
        57,  58,  59,  60,  67,  68,  69,  70,  71,  72,  79,  80,  81,
        82,  83,  84,  91,  92,  93,  94,  95,  96, 103, 104, 105, 106,
       107, 108, 115, 116, 117, 118, 119, 120], dtype='timedelta64[ns]')

Issue Number

Closes #2264

Is this PR a draft? Mark it as draft.

Checklist before asking for review

• I have performed a self-review of my code
• My changes comply with basic sanity checks:
  • I have fixed formatting issues with ./scripts/actions.sh lint
  • I have run unit tests with ./scripts/actions.sh unit-test
  • I have documented my code and I have updated the docstrings.
  • I have added unit tests, if relevant
• I have tried my changes with data and code:
  • I have run the integration tests with ./scripts/actions.sh integration-test
  • (bigger changes) I have run a full training and I have written in the comment the run_id(s): launch-slurm.py --time 60
  • (bigger changes and experiments) I have shared a hegdedoc in the github issue with all the configurations and runs for this experiments
• I have informed and aligned with people impacted by my change:
  • for config changes: the MatterMost channels and/or a design doc
  • for changes of dependencies: the MatterMost software development channel

[enssow]

Tested with both netcdf and verif parser using id=cae70ju8; s=ERA5; uv run export --run-id $id --stream $s --output-dir ../${id}_${s} --format verif --obs ../metno_observations_v3.nc and id=cae70ju8; s=ERA5; uv run export --run-id $id --stream $s --output-dir ../${id} --format netcdf on SANTIS; both produced expected results