Mutates Unicode encoding and byte-order specifiers in bitstring constructors. Each mutant changes one segment along one of these axes.
Encoding
A segment using utf8, utf16, or utf32 receives replacements using each of the other encodings. These encodings accept the same Unicode scalar values but produce different byte widths for non-empty values.
Byte order
big and little are exchanged for utf16 and utf32 segments. A bare UTF-16 or UTF-32 specifier defaults to big-endian and receives a variant with little added. UTF-8 has no byte-order mutation. native is not used as a source or replacement because its result depends on the host architecture.
For literal integer and binary values, each candidate is encoded before emission. A candidate whose bytes equal the original is removed. This filters empty values and byte-palindromic values whose byte order is unobservable. Variable values remain eligible because some runtime input can distinguish the encodings.
The family applies only to runtime bitstring constructors. Bitstring patterns are not mutated because an in-place selector cannot wrap a pattern. Interpolated strings and segments without a UTF encoding are also excluded. UTF segments with size or unit specifiers are not valid targets. Segment values may still be mutated independently by other families.