Report time dimension in geolocation output and add multi-time integration tests

[Copilot]

• Understand the existing implementation (time dim detection and nearest time selection already exist in engine.py)
• Update _geoloc_description() in _open_method.py to include time dim when present
• Update open_dataset() in plan.py to detect time dim and pass it to _geoloc_description()
• Add tests for time dim reporting in open_dataset()
• Add integration tests using multi_time.nc fixture
• Run code review and address feedback
• Run CodeQL security check
• Merge main (PR Rename spatial methods; fix opened_batch mutation bug; add xoak-haversine #135): spatial method renames (ndpoint→kdtree, xoak→xoak-kdtree, xoak-haversine added), opened_batch mutation fix

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Add an implementation of time method = "nearest" that finds the files that are closest in time.</issue_title>
<issue_description>Task: Implement finding nearest time if a xarray Dataset has time dim and multiple time points. If there is time dim in xarray dataset but only 1 time, then matching to nearest time is not necessary (skip).

Here is a sample netcdf with 366 days for testing
https://github.com/fish-pace/point-collocation/tree/main/examples/fixtures/multi_time.nc

Here is a notebook that shows some sample points and finding nearest lat/lon/time
https://github.com/fish-pace/point-collocation/blob/main/examples/multi_time.ipynb

Background: In some cases, a granule will have multiple times. For example, a yearly granule with 12 months or 365 days. A lat/lon/time will match to that granule, but the time needs to be matched to nearest time.

Notes:

• time should always be 1D. but might not be uniform.
• although unusual, it is possible that 2D lat/lon would have multiple times.
• if there is only 1 time, then matching to nearest time is not necessary
• cf_xarray should be used to find the time dim, and a reasonable fallback used.
• report time dim with geolocation in open_dataset() and anywhere else that geolocation is reported. Do not report time dim if the granules do not have a time dim</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Add an implementation of time method = "nearest" that finds the files that are closest in time. #43

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[eeholmes]

@copilot
I just did a merge into main
#135

It hopefully will not affect the code in the current PR. But review the changes from PR 135, pull into your branch and make any necessary updates.

[Copilot]

@copilot
I just did a merge into main
#135
...

Pulled in PR #135 cleanly — automatic merge, no conflicts. All 304 tests pass (commit 5e3ad4a). The changes from PR #135 (spatial method renames: ndpoint→kdtree, xoak→xoak-kdtree, new xoak-haversine, and the opened_batch mutation bugfix) don't touch _open_method.py or plan.py, so there's no overlap with this PR's changes.