[Plugin] Collect exact filesizes in add_dir_listing

[pmoravec]

add_dir_listing method in #3664 hid exact filesizes in "ls -l" like outputs in favour of human readable output. That has two drawbacks:

1. It provides less precise information, while particular filesizes for some files can be essential to know. E.g. 3.1K can mean 3075 or 3174 bytes and the difference can matter for some well known files with well known default sizes.
2. It makes it difficult to parse the size by various analytical tools

sos tool should be data collectio tool rather than interpreting them. So let replace human readable filesize by exact one.

Resolves: #3758

---

Please place an 'X' inside each '[]' to confirm you adhere to our Contributor Guidelines

• Is the commit message split over multiple lines and hard-wrapped at 72 characters?
• Is the subject and message clear and concise?
• Does the subject start with [plugin_name] if submitting a plugin patch or a [section_name] if part of the core sosreport code?
• Does the commit contain a Signed-off-by: First Lastname email@example.com?
• Are any related Issues or existing PRs properly referenced via a Closes (Issue) or Resolved (PR) line?
• Are all passwords or private data gathered by this PR obfuscated?

[packit-as-a-service]

Congratulations! One of the builds has completed. 🍾

You can install the built RPMs by following these steps:

• sudo yum install -y dnf-plugins-core on RHEL 8
• sudo dnf install -y dnf-plugins-core on Fedora
• dnf copr enable packit/sosreport-sos-3758
• And now you can install the packages.

Please note that the RPMs should be used only in a testing environment.

[TurboTurtle]

Should this be a toggle? When I standardized the existing calls to add_dir_listing() there was no clear preference across plugins as to if -h was used or not in the "manual" ls collections.

I'm also fine with just dropping human sizes across the board, just a thought.

[pmoravec]

Good point. Sometimes (e.g. https://github.com/sosreport/sos/pull/3664/files#diff-2e1b7ad69a0e01805218f011055c3ce7d123809531eed5ed157e004be5fe315bR175 or https://github.com/sosreport/sos/pull/3664/files#diff-b8035c7551e828c6ec017b8b2a986cce1884a2d87110a9edfa11bbdf4c40f561R57) the ls output is meant rather user-informative than precisely descriptive. But still I feel making the -h option toggle-able would "cost" us on code (execution) complexity more than users would gain. So until there would be a push back, I would stay with the -h drop.

I will just bump the PR to update the failing test.