ENH: Remove type hint in overriden diag methods

Remove type hint in overriden `diag` methods in kernel classes: it is
unsafe to override a method with a more specific argument type, as it
violates the Liskov substitution principle.

Fixes:
```
src/nifreeze/model/gpr.py:335: error:
 Argument 1 of "diag" is incompatible with supertype "Kernel";
 supertype defines the argument type as
 "Buffer | _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | bool | int | float | complex | str | bytes | _NestedSequence[bool | int | float | complex | str | bytes]"
  [override]
src/nifreeze/model/gpr.py:335: note: This violates the Liskov substitution principle
src/nifreeze/model/gpr.py:335: note: See https://mypy.readthedocs.io/en/stable/common_issues.html#incompatible-overrides
src/nifreeze/model/gpr.py:445: error:
 Argument 1 of "diag" is incompatible with supertype "Kernel";
 supertype defines the argument type as
 "Buffer | _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | bool | int | float | complex | str | bytes | _NestedSequence[bool | int | float | complex | str | bytes]"
  [override]
src/nifreeze/model/gpr.py:445: note: This violates the Liskov substitution principle
src/nifreeze/model/gpr.py:335: note: See https://mypy.readthedocs.io/en/stable/common_issues.html#incompatible-overrides
```

raised for example in:
https://github.com/nipreps/nifreeze/actions/runs/12437972140/job/34728973936#step:8:93

Documentation:
https://mypy.readthedocs.io/en/stable/common_issues.html#incompatible-overrides

Author: jhlegarreta

src/nifreeze/model/gpr.py

@@ -332,7 +332,7 @@ def __call__(
 
         return self.beta_l * C_theta, K_gradient
 
-    def diag(self, X: np.ndarray) -> np.ndarray:
+    def diag(self, X) -> np.ndarray:
         """Returns the diagonal of the kernel k(X, X).
 
         The result of this method is identical to np.diag(self(X)); however,
@@ -442,7 +442,7 @@ def __call__(
 
         return self.beta_l * C_theta, K_gradient
 
-    def diag(self, X: np.ndarray) -> np.ndarray:
+    def diag(self, X) -> np.ndarray:
         """Returns the diagonal of the kernel k(X, X).
 
         The result of this method is identical to np.diag(self(X)); however,