Ruff check

[mabruzzo]

EDIT: the contributions of this PR are now a part of #434. We can close this PR if we want

---

Overview

This adds the ruff linter to pre-commit.

Motivation

In more detail, the ruff linter performs similar actions to clang-tidy, but applies to python.

Important

Ruff also provides a code-formatter (e.g. like clang-format for python). This PR does NOT enable the formatter. While I think it would be useful to enable the code formatter, that's beyond the scope of this PR (it would produce a bunch of churn in the existing scripts and it provides less value)

It's useful to draw comparisons to clang-tidy:

• I think most of us share the sentiment that clang-tidy provides us with only limited benefit. In practice, it mostly just enforces coding conventions and infrequently catches bugs¹
• while the ruff linter also enforces coding-conventions, it serves some additional purposes. To illustrate this, consider the following (contrived) code block:

  a = 3
  if cond:
      a = a + 2
  else:
      a = a + myvar

  Now that this is part of a larger function where I forgot to define the variable myvar (or I defined it as my_var). Python won't identify the error unless we run the code and cond has a value of False. The ruff linter provides a lot of additional value² by identifying these cases for us!

The second bullet point covers the majority of python bugs that I personally produce (and that I see). And for that reason I think the linter is extremely useful (I found it useful while writing #427)

Details

Currently the ruff linter identifies lots of "problems" with our existing scripts. While many of these problems are about coding-style, some of them are real issues. To avoid making this a large PR, I configured the ruff linter to exclude all existing files and only apply to newly introduced files. But, we can fix that going forward.

As a proof of concept, I did address the "problems" in 2 existing files -- pertaining to unused import-statements

Footnotes

1. I suspect that this has a lot to do with the kind of logic within Cholla. I suspect that in more-general codebases (not just a numerical simulation code) it's probably a more useful tool. ↩

2. If this logic were in C/C++, clang-tidy could probably also identify this case for us. But this isn't as useful since any modern C/C++ compiler should also identify the error. ↩

[mabruzzo]

The changes introduced here actually got merged in with the cholla_utils PR. Consequently, I'm going to close it