[QianyeSu]

Related to #86.

cdsapi.api.Client.download(result, "download.nc") currently splits the filename into characters, passing only "d" to the result downloader. A bytes path is split into integers, and a pathlib.Path raises TypeError because it is not iterable.

Treat a single string, bytes or os.PathLike target as one path. Keep iterable targets copied before consumption and preserve existing list/tuple and default-filename behavior. Add a method docstring and offline regression tests, including file writes through mocked HTTP responses.

This change is scoped to cdsapi.api.Client. Modern token-based clients delegate to the separate ecmwf-datastores-client implementation, which is not modified here.

Validation on Python 3.14:

• pytest -q tests/test_download.py: 25 passed.
• Existing instantiation tests plus the new tests, using dummy credentials with HTTP requests blocked: 29 passed, 1 live retrieval test deselected.
• Black, isort, flake8, compile checks and git diff --check: passed.

No real CDS credentials or data downloads were used.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.