seq: add f128 code paths

[jalil-salame]

• refactors uumain

• adds regression test for seq 4e4000003 4e4000003 is causing an infinite loop #6182 (commented out until implementation is done)

• preliminary implementation of the f128 code path (feature gated)

  I'm having issues with libquadmath.so.0: it loads fine when I run the binary directly (target/debug/coreutils seq ...), but fails when I run the tests (cargo test/cargo nextest run). It might be a weird NixOS issue though.

I'll try using the nightly f128 code tomorrow to verify the tests pass.

[github-actions]

GNU testsuite comparison:

Skipping an intermittent issue tests/tail/inotify-dir-recreate (passes in this run but fails in the 'main' branch)

[jalil-salame]

The main features this needs to work on nightly are impl FromStr for f128 and impl Display for f128.

I tried implementing them by converting to/from BigDecimal, but that is a lot of code (hard to maintain) and I don't think I even did it correctly as a bunch of tests are failing.

I'll keep this as a draft and keep an eye on the f128 stabilization process; whenever FromStr and Display are stabilized I'll give it another try.

If anyone in the future wants to give this a shot, then check the following links:

• Docs on f128: https://doc.rust-lang.org/stable/std/primitive.f128.html
• f128 impl tracking issue: Tracking Issue for f16 and f128 float types rust-lang/rust#116909
• Wikipedia on IEEE f128: https://en.wikipedia.org/wiki/Quadruple-precision_floating-point_format

[sylvestre]

@jalil-salame it is still marked as draft, are you still working on it? thanks

[jalil-salame]

The main features this needs to work on nightly are impl FromStr for f128 and impl Display for f128.

I tried implementing them by converting to/from BigDecimal, but that is a lot of code (hard to maintain) and I don't think I even did it correctly as a bunch of tests are failing.

I'll keep this as a draft and keep an eye on the f128 stabilization process; whenever FromStr and Display are stabilized I'll give it another try.

If anyone in the future wants to give this a shot, then check the following links:

• Docs on f128: https://doc.rust-lang.org/stable/std/primitive.f128.html
• f128 impl tracking issue: Tracking Issue for f16 and f128 float types rust-lang/rust#116909
• Wikipedia on IEEE f128: https://en.wikipedia.org/wiki/Quadruple-precision_floating-point_format

It is blocked by these issues, I left it open so that contributors could see this before starting an attempt, if you feel it should be closed/reattempted, feel free to do so