Use UTF-8 for test case JSON files

[Ghraven]

Problem

A few Burr test-case JSON helpers still rely on the platform default text encoding when reading or writing JSON/JSONL files. That can make generated test fixtures or lifecycle logs less portable on systems where the default encoding is not UTF-8.

Before / after

Before, burr test-case create, load_test_cases(), and StateAndResultsFullLogger opened text JSON files without an explicit encoding.

After, those reads/writes use encoding="utf-8", matching the rest of Burr's text-file handling and keeping non-ASCII fixture/log content deterministic across platforms.

Verification

• python -m py_compile burr\cli_main_.py burr\lifecycle\default.py burr\testing_init_.py
• Ran a small local smoke check that writes/loads a UTF-8 JSON test case and writes UTF-8 JSONL content through StateAndResultsFullLogger.