Run Ruff on Markdown code blocks

[evanrittenhouse]

Summary

Closes #3792.

This is a draft PR tracking the development of Ruff's support for Markdown. I'm opening it now to get feedback from maintainers on if the proposed design makes sense.

The next step will involve changing lint_path and subsequent calls to work over str slices. Python/Jupyter notebooks will always be slices of length 1 - Markdown code blocks will be variable length. We can't concatenate the Markdown source code since the code blocks are independent, unlike Jupyter code blocks. I'd like to get the design solidified before I undertake that work.

There are still some abstractions that could be made, like converting our Jupyter implementation's mapping over to a CodeBlock-based one, as well as standardizing an interface for parsing files that don't solely contain Python. Same as above - I'd like to align on an initial CodeBlock design first, since that will likely be foundational to future embedded-language tasks.

Test Plan

[github-actions]

PR Check Results

Ecosystem

✅ ecosystem check detected no changes.

Benchmark

Linux

group                                      main                                   pr
-----                                      ----                                   --
formatter/large/dataset.py                 1.00      8.4±0.23ms     4.8 MB/sec    1.01      8.5±0.27ms     4.8 MB/sec
formatter/numpy/ctypeslib.py               1.02      2.0±0.07ms     8.3 MB/sec    1.00  1975.1±51.02µs     8.4 MB/sec
formatter/numpy/globals.py                 1.05   251.8±28.61µs    11.7 MB/sec    1.00    239.1±7.52µs    12.3 MB/sec
formatter/pydantic/types.py                1.01      4.5±0.13ms     5.7 MB/sec    1.00      4.4±0.10ms     5.8 MB/sec
linter/all-rules/large/dataset.py          1.00     17.1±0.44ms     2.4 MB/sec    1.00     17.1±0.45ms     2.4 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.00      4.2±0.16ms     4.0 MB/sec    1.00      4.2±0.12ms     4.0 MB/sec
linter/all-rules/numpy/globals.py          1.04   565.6±18.78µs     5.2 MB/sec    1.00   541.8±15.75µs     5.4 MB/sec
linter/all-rules/pydantic/types.py         1.01      7.6±0.20ms     3.4 MB/sec    1.00      7.5±0.18ms     3.4 MB/sec
linter/default-rules/large/dataset.py      1.02      8.5±0.14ms     4.8 MB/sec    1.00      8.4±0.21ms     4.8 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.00  1816.8±37.48µs     9.2 MB/sec    1.00  1815.6±55.72µs     9.2 MB/sec
linter/default-rules/numpy/globals.py      1.00    216.3±6.97µs    13.6 MB/sec    1.02   221.2±11.37µs    13.3 MB/sec
linter/default-rules/pydantic/types.py     1.00      3.9±0.08ms     6.6 MB/sec    1.02      4.0±0.19ms     6.4 MB/sec

Windows

group                                      main                                   pr
-----                                      ----                                   --
formatter/large/dataset.py                 1.00      8.0±0.14ms     5.1 MB/sec    1.01      8.1±0.12ms     5.0 MB/sec
formatter/numpy/ctypeslib.py               1.01  1856.8±53.15µs     9.0 MB/sec    1.00  1832.8±33.08µs     9.1 MB/sec
formatter/numpy/globals.py                 1.01    224.2±7.09µs    13.2 MB/sec    1.00   222.2±11.35µs    13.3 MB/sec
formatter/pydantic/types.py                1.00      4.1±0.06ms     6.2 MB/sec    1.01      4.1±0.08ms     6.2 MB/sec
linter/all-rules/large/dataset.py          1.01     16.0±0.23ms     2.5 MB/sec    1.00     15.9±0.30ms     2.6 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.00      4.2±0.06ms     4.0 MB/sec    1.00      4.2±0.10ms     4.0 MB/sec
linter/all-rules/numpy/globals.py          1.00    509.1±9.10µs     5.8 MB/sec    1.07  545.6±156.86µs     5.4 MB/sec
linter/all-rules/pydantic/types.py         1.02      7.1±0.15ms     3.6 MB/sec    1.00      7.0±0.14ms     3.7 MB/sec
linter/default-rules/large/dataset.py      1.00      8.4±0.13ms     4.9 MB/sec    1.06      8.9±0.67ms     4.6 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.00  1778.1±27.74µs     9.4 MB/sec    1.09  1943.8±711.33µs     8.6 MB/sec
linter/default-rules/numpy/globals.py      1.00    203.8±4.76µs    14.5 MB/sec    1.01    205.1±4.30µs    14.4 MB/sec
linter/default-rules/pydantic/types.py     1.00      3.7±0.05ms     6.9 MB/sec    1.01      3.8±0.06ms     6.8 MB/sec

[evanrittenhouse]

Tagging @MichaReiser for design review - no rush on it as Charlie said it's deprioritized, but it's a fun project that I enjoy working on

This is obviously incredibly rough right now, currently looking to see if the CodeBlock design is solid as it'll be used as the foundational unit for linting rather than a Path