fix typos in docstrings and docu.

docs/source/algorithms.md

@@ -392,7 +392,7 @@ install optimagic.
     .. warning::
         In our benchmark using a quadratic objective function, the trust_constr
         algorithm did not find the optimum very precisely (less than 4 decimal places).
-        If you require high precision, you should refine an optimum found with Powell
+        If you require high precision, you should refine an optimum found with trust_constr
         with another local optimizer.
 
     .. note::

src/optimagic/optimizers/_pounders/gqtpar.py

@@ -55,7 +55,7 @@ def gqtpar(model, x_candidate, *, k_easy=0.1, k_hard=0.2, maxiter=200):
             - ``linear_terms``, a np.ndarray of shape (n,) and
             - ``square_terms``, a np.ndarray of shape (n,n).
         x_candidate (np.ndarray): Initial guess for the solution of the subproblem.
-        k_easy (float): topping criterion for the "easy" case.
+        k_easy (float): Stopping criterion for the "easy" case.
         k_hard (float): Stopping criterion for the "hard" case.
         maxiter (int): Maximum number of iterations to perform. If reached,
             terminate.

src/optimagic/optimizers/_pounders/pounders_auxiliary.py

@@ -240,7 +240,7 @@ def solve_subproblem(
         gtol_rel_conjugate_gradient (float): Convergence tolerance for the relative
             gradient norm in the conjugate gradient step of the trust-region
             subproblem ("bntr").
-        k_easy (float): topping criterion for the "easy" case in the trust-region
+        k_easy (float): Stopping criterion for the "easy" case in the trust-region
             subproblem ("gqtpar").
         k_hard (float): Stopping criterion for the "hard" case in the trust-region
             subproblem ("gqtpar").

src/optimagic/optimizers/pounders.py

@@ -262,7 +262,7 @@ def internal_solve_pounders(
         gtol_rel_conjugate_gradient_sub (float): Convergence tolerance for the
             relative gradient norm in the conjugate gradient step of the trust-region
             subproblem if "cg" is used as ``conjugate_gradient_method_sub`` ("bntr").
-        k_easy_sub (float): topping criterion for the "easy" case in the trust-region
+        k_easy_sub (float): Stopping criterion for the "easy" case in the trust-region
             subproblem ("gqtpar").
         k_hard_sub (float): Stopping criterion for the "hard" case in the trust-region
             subproblem ("gqtpar").