fix: handle UnicodeDecodeError in git ls-files output

[sentry]

This PR addresses a UnicodeDecodeError occurring in GitVersioningSystem.list_relevant_files when processing the output of git ls-files -z.

Previously, res.stdout.decode() was used without an error handler, causing the CLI to crash if git ls-files -z returned filenames containing byte sequences that were not valid UTF-8. This can happen with filesystems or repositories where filenames use non-UTF-8 encodings (e.g., Latin-1).

The fix modifies the decode() call to res.stdout.decode(errors="replace"). This ensures that any invalid byte sequences encountered during decoding are replaced with the Unicode replacement character (\ufffd) instead of raising an error. This makes the CLI more robust and prevents crashes when encountering such filenames, allowing the upload process to complete successfully.

Fixes CLI-AK

_{This PR was automatically generated by Sentry. You can adjust this setting at any time.}

[codecov]

❌ 1 Tests Failed:

Tests completed	Failed	Passed	Skipped
723	1	722	1

View the full list of 1 ❄️ flaky test(s)

api.temp.calculator.test_calculator::test_divide

Flake rate in main: 100.00% (Passed 0 times, Failed 1726 times)

Stack Traces | 0.001s run time

def
                test_divide():
                > assert Calculator.divide(1, 2) == 0.5
                E assert 1.0 == 0.5
                E + where 1.0 = <function Calculator.divide at 0x104c9eb90>(1, 2)
                E + where <function Calculator.divide at 0x104c9eb90> = Calculator.divide
                .../temp/calculator/test_calculator.py:30: AssertionError

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}