tail: don't panic on '-c +N' with a huge N

[nagendramohan]

tail -c +N FILE on a regular file larger than the block size seeks to byte N-1. When N-1 exceeds what the OS can seek to (any offset above i64::MAX always fails with EINVAL), the .unwrap() aborted the process.

Clamp the start offset to the file length. A start beyond the end of the file produces no output (matching GNU tail), and clamping also avoids seeking to an offset the OS cannot represent — so the pathological case yields empty output instead of panicking. Behavior is unchanged for valid offsets.

Adds a regression test (verified failing before the fix). Closes #13887.

Closes #13887.

tail -c +N FILE on a regular file larger than the block size seeks to byte N-1. When N-1 exceeds what the OS can seek to (any offset above i64::MAX fails with EINVAL), the .unwrap() aborted the process:

$ yes | head -c 8192 > big
$ tail -c +18446744073709551615 big # -> SIGABRT / exit 134

Fix: clamp the start offset to the file length. A start beyond EOF produces no output (matching GNU tail), and clamping avoids the un-seekable offset
entirely. Behavior is unchanged for valid offsets.

Adds a regression test (verified failing before the fix). cargo fmt, cargo clippy, and the tail test suite pass.

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/tail/end-of-device. tests/tail/end-of-device is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/tail/symlink (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/tail/tail-n0f (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/misc/io-errors (passes in this run but fails in the 'main' branch)