fix(hermes_time): read config yaml as utf-8

[acefun29]

What does this PR do?

Fixes timezone config loading when config.yaml is UTF-8 but the platform default text encoding is different.

Previously, hermes_time.py opened the user's config.yaml without an explicit encoding. On systems where the default encoding is not UTF-8, a config file containing non-ASCII UTF-8 text could fail to decode and silently fall back to the local timezone. This PR reads the config as UTF-8 explicitly, matching how Hermes config files are normally written.

Related Issue

N/A

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)
• ✅ Tests (adding or improving test coverage)

Changes Made

• Updated hermes_time.py to read config.yaml with encoding="utf-8".
• Added a regression test in tests/test_timezone.py that simulates a non-UTF-8 default encoding while reading a UTF-8 config file.

How to Test

python -m pytest tests/test_timezone.py -q -o addopts=

Result:

19 passed, 2 skipped

Checklist

• My commit messages follow Conventional Commits
• My PR contains only changes related to this fix
• I've added tests for my changes
• I've tested on Windows 11, Python 3.11.9

[alt-glitch]

Likely duplicate of #8455 — same UTF-8 encoding fix for hermes_time.py config read. Part of broader #15519 effort.

[alt-glitch]

Likely duplicate of #8455 — same UTF-8 encoding fix for hermes_time.py config read.

[teknium1]

Closing as already implemented on main by the broader UTF-8 file-open sweep.

Evidence from this automated hermes-sweeper review:

• hermes_time.py:53 now reads the config path with encoding="utf-8", which is the production-code change requested here.
• Commit cbce5e93fcb9a923ab71f45d2a0f0f172dd54967 made that change and describes the Windows locale-decoding bug class this PR targets.
• The fix is contained in release tag v2026.5.16 and later release tags.
• The existing maintainer discussion on this PR already identified it as a duplicate of fix(time): read timezone from config.yaml as UTF-8 #8455 / the broader fix: add explicit encoding="utf-8" to open() calls for cross-platform safety #15519 UTF-8 encoding effort.

Thanks for the focused Windows repro and regression-test contribution.