Linter: Update to ruff 0.15.0

[amotl]

Problem

This unrelated patch just failed CI because the ruff linter complained.

• CI: Validate on free-threaded Python (GHA label 3.14t) #773

Thoughts

I don't know why the CI run on that update by Dependabot succeeded. I think the CI run on that update by Dependabot didn't run yesterday, because it was submitted at a time when GitHub had outages across the board.

• Chore(deps-dev): Update ruff requirement from <0.15 to <0.16 #775

[amotl]

Without that marker, ruff 0.15.0 would complain like this:

A003 Python builtin is shadowed by method `next` from line 181
   --> src/crate/client/cursor.py:196:16
    |
194 |             raise ProgrammingError("Cursor closed")
195 |
196 |     __next__ = next
    |                ^^^^
197 |
198 |     @property
    |

Found 1 error.

[amotl]

Would you suggest a different mitigation than just suppressing the warning?

[amotl]

Is it the right choice to use builtins.next?

Suggested change

	__next__ = next # noqa: A003
	__next__ = builtins.next

[mfussenegger]

Judging from 1d7573e the intention here is to implement the iterator protocol

So why not:

diff --git a/src/crate/client/cursor.py b/src/crate/client/cursor.py
index 587f149..b7aac08 100644
--- a/src/crate/client/cursor.py
+++ b/src/crate/client/cursor.py
@@ -191,11 +191,12 @@ def next(self):
         if not self._closed:
             return next(self.rows)
         else:
             raise ProgrammingError("Cursor closed")
 
-    __next__ = next
+    def __next__(self):
+        return self.next()
 
     @property
     def description(self):
         """
         This read-only attribute is a sequence of 7-item sequences.

[mfussenegger]

See comment regarding the next - otherwise lgtm