[release] rephrase some changes

CHANGELOG.md

@@ -90,32 +90,32 @@ Supported systems and requirements:
 + Add (S)SOR and Gauss-Seidel preconditioner [#1633](https://github.com/ginkgo-project/ginkgo/pull/1633), [#1634](https://github.com/ginkgo-project/ginkgo/pull/1634)
 + Add support for additive `read_distributed` for the distriubted matrix [#1650](https://github.com/ginkgo-project/ginkgo/pull/1650)
 + Add Ginkgo's own ILU and IC implementation [#1684](https://github.com/ginkgo-project/ginkgo/pull/1684)
-+ Add Nvidia ada architecture [#1733](https://github.com/ginkgo-project/ginkgo/pull/1733)
++ Add NVIDIA ada architecture [#1733](https://github.com/ginkgo-project/ginkgo/pull/1733)
 + Add half precision support [#1706](https://github.com/ginkgo-project/ginkgo/pull/1706), [#1708](https://github.com/ginkgo-project/ginkgo/pull/1708), [#1711](https://github.com/ginkgo-project/ginkgo/pull/1711), [#1712](https://github.com/ginkgo-project/ginkgo/pull/1712), [#1713](https://github.com/ginkgo-project/ginkgo/pull/1713), [#1716](https://github.com/ginkgo-project/ginkgo/pull/1716), [#1710](https://github.com/ginkgo-project/ginkgo/pull/1710), [#1736](https://github.com/ginkgo-project/ginkgo/pull/1736)
 
 ### Improvements
 + Add workspace in residual norm check [#1687](https://github.com/ginkgo-project/ginkgo/pull/1687), which reduces the alloc/free and corresponding overhead.
 + Add distributed `VectorCache` and use it as workspace in `Schwarz` [#1688](https://github.com/ginkgo-project/ginkgo/pull/1688).
 + Add example to show the file config usage [#1662](https://github.com/ginkgo-project/ginkgo/pull/1662)
 + Improve compile time for batched solvers [#1629](https://github.com/ginkgo-project/ginkgo/pull/1629)
-+ Reduce the thrust conflict possibility by adding custom thrust namespace [#1730](https://github.com/ginkgo-project/ginkgo/pull/1730)
++ Reduce conflicting thrust symbols when linking with different thrust libraries by adding a custom thrust namespace [#1730](https://github.com/ginkgo-project/ginkgo/pull/1730)
 
 ### Fixes
-+ Fix the algorithm usage of transposed triangular solver, which is the same as the original one not the default algorithm [#1641](https://github.com/ginkgo-project/ginkgo/pull/1641)
++ Fix using the same algorithm as the original triangular solver when creating the transposed of the solver [#1641](https://github.com/ginkgo-project/ginkgo/pull/1641)
 + Fix the inconsistent behavior on the zero diagonal value in scalar Jacobi [#1642](https://github.com/ginkgo-project/ginkgo/pull/1642)
 + Fix an issue related to GCR and non-default strides in the rhs vector [#1656](https://github.com/ginkgo-project/ginkgo/pull/1656)
 + Fix an issue related to triangular solvers with CUDA on Windows [#1665](https://github.com/ginkgo-project/ginkgo/pull/1665)
 + Fix an issue where non-conforming MatrixMarket files were parsed without an error [#1628](https://github.com/ginkgo-project/ginkgo/pull/1628)
 + Fix finding rocthrust if it's not installed paths included by default [#1668](https://github.com/ginkgo-project/ginkgo/pull/1668)
-+ Fix the casting issue in mixed-precision multigrid setup [#1663](https://github.com/ginkgo-project/ginkgo/pull/1663)
++ Fix an issue related to casting between vectors of different value types in the mixed-precision multigrid setup [#1663](https://github.com/ginkgo-project/ginkgo/pull/1663)
 + Fix some test failures with ROCm 6.x [#1670](https://github.com/ginkgo-project/ginkgo/pull/1670)
-+ Fix the race condition in bicgstab [#1676](https://github.com/ginkgo-project/ginkgo/pull/1676)
++ Fix a race condition in bicgstab [#1676](https://github.com/ginkgo-project/ginkgo/pull/1676)
 + Fix an issue with MGS GMRES for complex numbers [#1678](https://github.com/ginkgo-project/ginkgo/pull/1678)
 + Fix finding ROCm on recent ROCm version (5.0+) [#1673](https://github.com/ginkgo-project/ginkgo/pull/1673)
-+ Fix compiler error when using NVHPC with MPI enabled [#1697](https://github.com/ginkgo-project/ginkgo/pull/1697)
-+ Fix build issue of OMP backend when using HIPCC as C++ compiler [#1695](https://github.com/ginkgo-project/ginkgo/pull/1695)
-+ Fix build issue for Intel OneAPI 2025.0 [#1718](https://github.com/ginkgo-project/ginkgo/pull/1718)
-+ Fix inconsitent function and class/struct, which mainly fixes clang-cl [#1725](https://github.com/ginkgo-project/ginkgo/pull/1725)
++ Fix a compiler error when using NVHPC with MPI enabled [#1697](https://github.com/ginkgo-project/ginkgo/pull/1697)
++ Fix build issues of OMP backend when using HIPCC as C++ compiler [#1695](https://github.com/ginkgo-project/ginkgo/pull/1695)
++ Fix build issues for Intel OneAPI 2025.0 [#1718](https://github.com/ginkgo-project/ginkgo/pull/1718)
++ Fix inconsistencies between declaration and definition of functions and classes/structs, which mainly fixes clang-cl [#1725](https://github.com/ginkgo-project/ginkgo/pull/1725)
 + Fix undefined symbols in shared library in msys2/clang [#1724](https://github.com/ginkgo-project/ginkgo/pull/1724)
 + Fix page fault issues when running on multiple Intel GPUs in parallel [#1723](https://github.com/ginkgo-project/ginkgo/pull/1723)