chore: release v0.1.139

CI in [1] seems to indicate that there are cases where the `f16`
infinite recursion bug ([2], [3]) can make its way into what gets called
during tests, even though this doesn't seem to be the usual case. In
order to make sure that we avoid these completely, just unset
`f16_enabled` on any platforms that have the recursion problem.

This also refactors the `match` statement to be more in line with
`library/std/build.rs`.

[1]: #729
[2]: llvm/llvm-project#97981
[3]: #651

Disable `f16` on platforms that have recursion problems

These implementations of `abs_diff` were added in c2ff1b3
("Completely overhaul fuzz testing"), but the signed implementation is
wrong when |x| + |y| exceeds the integer's limits (e.g.
`(-128).abs_diff(1)` should be 128 but currently these return 127.

Resolve this by just using `std`'s implementation since that is stable
now. This isn't used anywhere critical, we probably just weren't hitting
the edge case.

Fix a bug in `abs_diff`

`compiler_builtins` fails to compile to amdgpu if f128 is enabled.
The reason seems to be that compiler_builtins uses libcalls in the
implementation. I’m not really familiar with what libcalls are, but the
LLVM amdgpu backend explicitly does not support them.

Error message:
```
LLVM ERROR: unsupported libcall legalization
```