Enable search and data file access by passing Dataset ID for PangaeaDataset

[doswal]

This PR adds support for direct loading of PANGAEA datasets using Study IDs (numeric IDs or DOI strings), bringing the PANGAEA provider closer to the behavior of the NOAA provider.

Previously, PangaeaDataset required users to:

• Run search_studies() with a query
• Inspect the returned summary
• Select a StudyID
• Call get_data() to retrieve the dataset

Unlike NOAA, PANGAEA could not directly load a dataset from its StudyID without a prior search.

What’s New

This PR introduces:

search_studies(study_ids=...)

Users can now load datasets directly using:

ds.search_studies(study_ids=830586)
ds.search_studies(study_ids="10.1594/PANGAEA.830586")
ds.search_studies(study_ids=[830586, "830587"])

• Supports single ID or list
• Accepts int or DOI string
• Automatically normalizes IDs
• Registers studies directly without invoking PanQuery

get_data(study_id) Auto-Loading

get_data() now supports:

ds.get_data(830586)
ds.get_data([830586, 830587])

• If study is not registered → it is loaded automatically
• If study is a collection → warning is logged and skipped

Always returns a list of pandas DataFrames (consistent with NOAA behavior)

The get_data related tests have also been modified to support this workflow