Fix to_datetime nanosecond inference and DatetimeIndex tz wall-time interpretation

[galipremsagar]

• to_datetime infers nanosecond precision for strings with more than 6 fractional-second digits, matching pandas.
• DatetimeIndex with a timezone dtype treats timezone-naive strings/datetimes as wall time in the target timezone, while numeric data stays interpreted as UTC epoch values.
• Removes 13 now-passing entries from the pandas-testing plugin.
• Split out from Fix factorize, value_counts, and to_datetime pandas compatibility issues #23138.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[galipremsagar]

/okay to test d4e2eba

[coderabbitai]

📝 Walkthrough

Walkthrough

Modifies DatetimeIndex construction to interpret timezone-naive strings as wall time when given a tz-aware dtype, updates to_datetime's strptime precision inference based on fractional-second format width, adjusts pandas-testing-plugin xfail expectations accordingly, and adds corresponding tests.

Changes

Datetime precision and timezone handling

Layer / File(s)	Summary
Timezone-aware DatetimeIndex dtype casting python/cudf/cudf/core/index.py	DatetimeIndex.__init__ now casts non-numeric data to a datetime64[unit] dtype then applies tz_localize when given a tz-aware DatetimeTZDtype, instead of a direct astype.
strptime fractional-second precision inference python/cudf/cudf/core/tools/datetimes.py	_process_col infers nanosecond vs microsecond target_unit by inspecting the %f directive width in format when unit is None; copyright header updated.
Pandas-testing-plugin xfail expectation updates python/cudf/cudf/pandas/scripts/pandas-testing-plugin.py	Removes/updates expected-failure entries for datetime constructor, tz-order, concat, pivot, unique/factorize, and nanosecond-parsing test cases.
New tests for precision inference and wall-time tz behavior python/cudf/cudf/tests/general_functions/test_to_datetime.py	Adds tests validating nanosecond/microsecond precision inference and wall-time interpretation for tz-aware DatetimeIndex; copyright header updated.

Estimated code review effort: 3 (Moderate) | ~25 minutes

Possibly related PRs

• rapidsai/cudf#22904: Both PRs modify python/cudf/cudf/pandas/scripts/pandas-testing-plugin.py by pruning/updating stale NODEIDS_THAT_* entries—main PR focuses on datetime/timezone-related xfails, while the retrieved PR removes general stale xfails and adjusts related node IDs.

Suggested labels: improvement, 3 - Ready for Review

Suggested reviewers: TomAugspurger, mroeschke, vyasr

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title clearly summarizes the two main datetime behavior fixes in the changeset.
Description check	✅ Passed	The description directly matches the implemented to_datetime, DatetimeIndex, and test plugin updates.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@python/cudf/cudf/tests/general_functions/test_to_datetime.py`:
- Around line 421-430: The test `test_datetime_index_tz_dtype_wall_time` only
covers string input for `cudf.DatetimeIndex` with `pd.DatetimeTZDtype("s",
"US/Eastern")`; add an integer epoch case in the same test so the numeric
construction path is also verified. Use a numeric timestamp input and assert it
matches the corresponding `pd.DatetimeIndex` result, ensuring `assert_eq` covers
both string and int behavior for this wall-time timezone dtype.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: ecaf5e04-21ab-4466-a196-e3306cf814a1

📥 Commits

Reviewing files that changed from the base of the PR and between ba255bb and d4e2eba.

📒 Files selected for processing (4)

• python/cudf/cudf/core/index.py
• python/cudf/cudf/core/tools/datetimes.py
• python/cudf/cudf/pandas/scripts/pandas-testing-plugin.py
• python/cudf/cudf/tests/general_functions/test_to_datetime.py

💤 Files with no reviewable changes (1)

• python/cudf/cudf/pandas/scripts/pandas-testing-plugin.py

[galipremsagar]

/okay to test f3915e7

[galipremsagar]

/merge