Add mimetype detection to walkfs

[JSCU-CNI]

As promised in #1387, this PR adds mimetype detection to the walkfs plugin.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 80.54%. Comparing base (eef181a) to head (740d759).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1521      +/-   ##
==========================================
- Coverage   80.80%   80.54%   -0.26%     
==========================================
  Files         396      396              
  Lines       34826    34836      +10     
==========================================
- Hits        28141    28060      -81     
- Misses       6685     6776      +91     

Flag	Coverage Δ
unittests	80.54% <100.00%> (-0.26%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[codspeed-hq]

Merging this PR will degrade performance by 82.42%

❌ 1 regressed benchmark
✅ 10 untouched benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
❌	test_benchmark_walkfs	58.9 ms	335.3 ms	-82.42%

---

_{Comparing JSCU-CNI:walkfs-magic (740d759) with main (eef181a)}

[Schamper]

I'm surprised this doesn't alter the benchmark. Can you say anything if it has a noticeable performance impact?

[JSCU-CNI]

I'm surprised this doesn't alter the benchmark.

The benchmark unfortunately only calls next() once. That means the benchmark only tests parsing the root filesystem entry.

Can you say anything if it has a noticeable performance impact?

Seems to be slightly slower in the test below on a ext4 filesystem with a default install of Ubuntu server (~100k files, ~4GB vmdk):

# amd64, cpython3.13

# without mimetype
$ time target-query -f walkfs ubuntu-server-24.04.2.vmx
...
real    13m3.404s
user    12m59.111s
sys     0m3.242s

# with mimetype
$ time target-query -f walkfs ubuntu-server-24.04.2.vmx
...
real    13m48.187s
user    13m30.432s
sys     0m5.915s

[B0TAxy]

Maybe we can make the MIME type generation optional? That way, we can switch it on or off as needed.

[Schamper]

The benchmark unfortunately only calls next() once.

Maybe change it to use itertools.islice with a sane number?

Maybe we can make the MIME type generation optional? That way, we can switch it on or off as needed.

I was thinking this too. Maybe use argparse.BooleanOptionalAction for a --mimetype/--no-mimetype flag.

[Schamper]

I know this record docstring is still a bit useless at this point, but would you mind moving it up to the right spot?

[Schamper]

I'm leaning towards flipping this and making it --mimetype to enable mimetype lookup. Sounds like a safer default?