Merge idx based masking with global/local views and ModelBatch

[shmh40]

Describe the task. It can be a feature, documentation, etc.

Combine changes with global/local views and ModelBatch/SampleMetadata with the index based masking. Requires adding production of global and local views (with healpix cropping or other strategies) to the index based masking which currently supports random masking, healpix on level of the data and forecast. Then piping the resulting StreamData through to the ModelBatch dataclass with associated metadata SampleMetadata in views.py in #1216 .

Combines some aspects of #1216 into #1242.

Hedgedoc URL, if you are keeping notes, plots, logs in hedgedoc.

No response

Area

• datasets, data readers, data preparation and transfer
• model
• science
• infrastructure and engineering
• evaluation, export and visualization
• documentation