Warning about "-9.000" -> NaN method in pandas 2

[zmoon]

uscrn/data.py:234: FutureWarning: Downcasting behavior in `replace` is deprecated and will be removed in a future version.
To retain the old behavior, explicitly call `result.infer_objects(copy=False)`.
To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`
  df["crx_vn"] = df["crx_vn"].replace("-9.000", np.nan)

[zmoon]

Seems like explicitly casting to desired dtype after replace may be the solution. But better identify the files this is happening for (seems like it may just be a few of the early archive ones) and check.

If it's that all version numbers are missing for a certain file, we don't actually want float dtype in that case.