spurious test failures `condvar::nonpoison_timeout_nanoseconds`

[matthiaskrgr]

happening since #146503 it seems x86_64-mingw-1 , x86_64-msvc-1
cc @joboet

#147508 (comment)
#147874 (comment)

2025-10-16T14:03:24.5908828Z failures:
2025-10-16T14:03:24.5909052Z 
2025-10-16T14:03:24.5909735Z ---- condvar::nonpoison_timeout_nanoseconds stdout ----
2025-10-16T14:03:24.5910121Z 
2025-10-16T14:03:24.5910856Z thread 'condvar::nonpoison_timeout_nanoseconds' (3328) panicked at library\std\tests\sync\condvar.rs:298:13:
2025-10-16T14:03:24.5912086Z assertion failed: *guard
2025-10-16T14:03:24.5912443Z stack backtrace:
2025-10-16T14:03:24.5912756Z    0: __rustc::rust_begin_unwind
2025-10-16T14:03:24.5913140Z    1: core::panicking::panic_fmt
2025-10-16T14:03:24.5913533Z    2: core::panicking::panic
2025-10-16T14:03:24.5913941Z    3: core::ops::function::FnOnce::call_once
2025-10-16T14:03:24.5914587Z note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.
2025-10-16T14:03:24.5915323Z ---- condvar::nonpoison_timeout_nanoseconds stdout end ----
2025-10-16T14:03:24.5915726Z 
2025-10-16T14:03:24.5915831Z failures:
2025-10-16T14:03:24.5916130Z     condvar::nonpoison_timeout_nanoseconds
2025-10-16T14:03:24.5916431Z 
2025-10-16T14:03:24.5916831Z test result: FAILED. 316 passed; 1 failed; 0 ignored; 0 measured; 0 filtered out;
``

[ChrisDenton]

#147871

[Zalathar]

#147897 (comment)