[BUG]: write_json / read_json raise UnicodeEncodeError on Windows for non-cp1252 characters

[LukeTheoJohnson]

Description

pio.write_json raises UnicodeEncodeError on Windows when a figure's text contains characters outside the cp1252 default codec. pio.read_json has the matching bug — it reads without specifying an encoding, so a UTF-8 JSON file is decoded with cp1252, mangling non-ASCII text or raising UnicodeDecodeError.

This works on macOS/Linux only because their default encoding is already UTF-8.

Steps to reproduce

import plotly.graph_objects as go
import plotly.io as pio

fig = go.Figure(layout_title_text="μ 中文")   # Greek mu + CJK
pio.write_json(fig, "fig.json")
# UnicodeEncodeError: 'charmap' codec can't encode character 'μ' in position ...: character maps to <undefined>

Environment: Windows, locale.getpreferredencoding(False) → cp1252, orjson installed.

Notes

Two conditions both have to hold:

1. The figure text contains a character outside cp1252. Common Latin-1 characters (é, ö, °) are inside cp1252 and do not raise — it takes Greek (μ), CJK, or emoji to trigger it.
2. orjson is installed (the default engine when available). orjson emits real UTF-8, so non-ASCII characters reach write_text unchanged. The pure-Python json engine uses ensure_ascii=True, escaping everything to \uXXXX ASCII, so it never trips the codec.

write_html already handles this correctly with encoding="utf-8" (see #3898). The JSON I/O path was missed.

[emilykl]

Addressed by #5633