Previously, the RISC-V write syscall in Ripes interpreted memory content
as a QString, which implicitly assumes UTF-8 encoding. This caused
incorrect output when writing binary data, particularly when bytes had
values above 127 (0x7F).

In UTF-8, codepoints above 127 are encoded using multiple bytes. For
example, the byte 0xA0 is encoded in UTF-8 as the two-byte sequence
0xC2 0xA0. As a result, using QString for raw memory meant that writing
a single byte from memory could result in multiple unintended bytes
being emitted to the output stream.

This patch replaces the use of QString with QByteArray, ensuring that
the written data is treated as a raw byte buffer rather than UTF-8
encoded text. This change preserves the exact memory contents intended
by the simulated program, enabling correct output through the write
syscall.