ls: implement device symbol and id

[tertsdiepraam]

Closes #2140

This is a quick fix to show the correct symbols for character device and block device, fifo and pipe special files and show the minor and major id numbers of devices. I'm not sure how to test this, so any ideas on that are welcome!

The id numbers are not spaced correctly yet, because that would require a bit of a rewrite of the long format. I would rather do that in another PR (which should also fix #2139). This is what it looks like now:

crw-rw---- 1 root audio 116, 12 Apr 29 09:01 hwC0D0
crw-rw---- 1 root audio 116, 11 Apr 29 09:01 hwC0D2
crw-rw---- 1 root audio  116, 5 Apr 29 13:08 pcmC0D0c
crw-rw---- 1 root audio  116, 4 Apr 29 13:08 pcmC0D0p

and here is what it should eventually look like:

crw-rw----+ 1 root audio 116, 12 29 apr 09:01 hwC0D0
crw-rw----+ 1 root audio 116, 11 29 apr 09:01 hwC0D2
crw-rw----+ 1 root audio 116,  5 29 apr 13:08 pcmC0D0c
crw-rw----+ 1 root audio 116,  4 29 apr 13:08 pcmC0D0p

Note: the + at the end is also not implemented yet.

[tertsdiepraam]

Looks like the birthtime tests are still not 100% stable...

[sylvestre]

@tertsdiepraam yeah, would you mind opening a bug about this?
(we can just ignore it)

[sylvestre]

Could you please fix

failures:

---- test_ls::test_ls_order_birthtime stdout ----
current_directory_resolved: 
open: /tmp/.tmpsgBoma/test-birthtime-1
run: /home/runner/work/coreutils/coreutils/target/debug/coreutils ls --time=birth -t
---- test_ls::test_ls_order_birthtime stderr ----
thread 'main' panicked at 'assertion failed: `(left == right)`
  left: `"test-birthtime-1\ntest-birthtime-2\n"`,
 right: `"test-birthtime-2\ntest-birthtime-1\n"`', /home/runner/work/coreutils/coreutils/tests/by-util/test_ls.rs:580:5
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace


failures:
    test_ls::test_ls_order_birthtime

[tertsdiepraam]

Seems like that other PR was not successful then :( I'll try again soon!

[sylvestre]

Could you please fix the conflict? thanks

[tertsdiepraam]

Should be all good now! Only code cov failed due to an unrelated test.

[sylvestre]

conflicting again with the recent pr :( sorry

[tertsdiepraam]

No problem! I was expecting that to happen :)

[sylvestre]

is it expected that we don't have tests for this?

[tertsdiepraam]

Yeah, I wasn't sure how to test that. If you know a way to create (or mock?) devices in the test suite I'd be happy to add some tests!

[sylvestre]

dunno, sorry