BUG: nanoseconds and reso in dateutil paths

[jbrockmendel]

• closes BUG: inferred Timestamp unit with dateutil paths #55737 (Replace xxxx with the GitHub issue number)
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

Perf impact is pretty negligible compared to the cost of going through dateutil:

import pandas as pd
import numpy as np

dtstr = "2016/01/02 03:04:05.001000 UTC"

%timeit pd.Timestamp(dtstr)
100 µs ± 3.05 µs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)  # <- PR
109 µs ± 10.1 µs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)  # <- main

vals = np.array([dtstr] * 10**5, dtype=object)
%timeit pd.to_datetime(vals)
8.37 ms ± 134 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <- PR
9.1 ms ± 597 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <- main

%timeit pd.to_datetime(vals, format="mixed")
8.29 ms ± 236 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <- PR
8.13 ms ± 63.6 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <- main

[mroeschke]

Are the first \ds always guaranteed to be separate by :?

[jbrockmendel]

dateutil supports some really weird formats (@MarcoGorelli and i have discussed moving away from using it at all) so i dont know. but i think this covers the vast majority of cases we care about

[mroeschke]

Thanks @jbrockmendel