fix(numfmt):fix recision loss for large numbers in #11654

[THEMIScale]

Fixes #11654

• Add numeric.rs to preserve precision for large integers
• Modify related functions in format.rs
• Adjust some float related test cases in format.rs
• Large integer values are now accurate
• Float precision remains limited by f64 and is worse than GNU
• cargo test --features unix passes for numfmt

[THEMIScale]

There's a remaining problem here:

┌──(kali㉿KaliOnHP)-[~/coreutils]
└─$ ./target/debug/numfmt --from=iec 915339622.7555392131
915339622.7555391788

┌──(kali㉿KaliOnHP)-[~/coreutils]
└─$ /usr/bin/numfmt --from=iec 915339622.7555392131
915339622.7555392131

The f64 is not as precise as long double.
Even to igore the case, GNU numfmt still has an error when number is out of limitation(which WE don't):

┌──(kali㉿KaliOnHP)-[~/coreutils]
└─$ ./target/debug/numfmt --from=iec 915339622.7555392723748798127937497128934
915339622.7555392980575561523437500000000

┌──(kali㉿KaliOnHP)-[~/coreutils]
└─$ /usr/bin/numfmt --from=iec 915339622.7555392723748798127937497128934
numfmt: 值或精度过大，无法输出："9.1534e+08/31"（请考虑使用 --to）

If anyone see this, pls NEW an ISSUE if it is reasonable, I'm worrying that this PR won't be merged, thus my environment is not relyable for now.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/pr/bounded-memory (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/cp/link-heap is now being skipped but was previously passing.
Congrats! The gnu test tests/expand/bounded-memory is now passing!