fix: preserve a leading BOM when set_key and unset_key rewrite a file

[MohammedAlkindi]

set_key and unset_key drop a leading UTF-8 BOM from files that have one.

#640 made the parser strip a leading BOM so the first variable parses. That stripped BOM never reaches the bindings, so the rewrite path writes the file back without it:

>>> pathlib.Path(".env").write_bytes(b"\xef\xbb\xbfa=x\n")
>>> set_key(".env", "b", "y")
>>> pathlib.Path(".env").read_bytes().startswith(b"\xef\xbb\xbf")
False

rewrite() now carries it across, so both writers keep it. Only files that already had a BOM change behaviour. The check is encoding-safe: a latin-1 file never decodes to \ufeff, and utf-16 handles its own BOM in the codec.

Tests: one for set_key, one for unset_key. Both fail on main on the missing BOM and pass with the fix. Full suite 207 passed / 50 skipped, ruff check and ruff format --check clean.

mypy reports one error here, tests/test_fifo_dotenv.py:15: Module has no attribute "mkfifo". It is pre-existing and unrelated: I confirmed it on a clean checkout of main. It appears because I ran mypy on Windows, where os.mkfifo does not exist, so CI will not see it.

Related: #640, #637