docs: remove numbers for mumps speedup, remove make_linear_sys

doc/source/whatsnew/0.2.rst

@@ -7,7 +7,7 @@ Improved performance
 --------------------
 This has been the main focus of this release.  Through optimization a level of
 performance has been reached that we consider satisfactory: runs of kwant for
-mid-sized (100x100 say) systems now typically spend most time in higly
+mid-sized (100x100 say) systems now typically spend most time in highly
 optimized libraries and not anymore in Python-implemented code.  For large,
 truly performance-critical systems almost all time is now spent in optimized
 libraries.
@@ -21,7 +21,7 @@ New MUMPS-based solver
 The code for sparse matrix solvers has been reorganized and a new solver has
 been added next to `kwant.solvers.sparse`: `kwant.solvers.mumps`.  The new
 solver uses the `MUMPS <http://graal.ens-lyon.fr/MUMPS/>`_ software package and
-is much (often 2-3 times and more) faster than the UMFPACK-based old solver.
+is much (typically several times) faster than the UMFPACK-based old solver.
 In addition, MUMPS uses considerably less memory for a given system while at
 the same time it is able to take advantage of more than 2 GiB of RAM.
 
@@ -89,7 +89,3 @@ The function `~kwant.solvers.common.SparseSolver.solve` of sparse solvers now
 always returns a single instance of `~kwant.solvers.common.BlockResult`.  The
 latter has been generalized to include more information for leads defined as
 infinite systems.
-
-Return value of `~kwant.solvers.common.SparseSolver.make_linear_sys` has changed
---------------------------------------------------------------------------------
-A namedtuple is used for more clarity.