install: retry strip program spawn on ETXTBSY

[krosci]

On Linux, exec(2) of a script (e.g. the #!/bin/sh strip wrapper) transiently fails with ETXTBSY when another process is concurrently exec'ing the same interpreter, because the kernel denies write access to the file for the duration of the exec; CI's parallel test load turns this into an intermittent failure. BSD kernels already sleep and retry on ETXTBSY (see exec(3)), and install now does the same, retrying the strip-program spawn up to three times with a short delay, so genuine errors (missing program, permission, non-zero exit) still fail immediately. Unit tests cover the retry path and confirm only ETXTBSY triggers a retry.

Closes #14520

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/date/date-locale-hour (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/cut/bounded-memory (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/date/resolution (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/tail/tail-n0f (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/tail/pipe-f is now being skipped but was previously passing.
Note: The gnu test tests/misc/write-errors was skipped on 'main' but is now failing.

[krosci]

@xtqqczze could you take a look at it? Thanks!

[xtqqczze]

could you take a look at it? Thanks!

With #14586, it would be possible to remove all the cfg(unix) annotations. I’ll wait for a decision on that PR before proceeding.