use autosetup for applying patches

it's error-prune to apply the patches using '%patch<n>' or '%patch'
macros. let's use %autosetup to do this job. as %autosetup defaults to
quiet operation, "-q" option is dropped.

also, in 28be73603a, we added this patch in hope to address a warning
from compiler. but in rawhide we use %autosetup macro to apply
the patches automatically, while in f38, we still use %patch<n>
to apply the patches one by one, so we failed to apply it when
cherry-picking the change from rawhide.

in this change, let's switch to autosetup for applying the patches.

this should apply patch7 to address the Boost.Accumulators bug filed against
boost package, see https://bugzilla.redhat.com/show_bug.cgi?id=2236519

this also silences the warning like

```
RPM build warnings:
    %patchN is deprecated (14 usages found), use %patch N (or %patch -P N)
```

see 6b80e21fd9 for the change on
rawhide

Signed-off-by: Kefu Chai <tchaikov@gmail.com>

boost-1.58.0-pool.patch

@@ -1,7 +1,7 @@
 Index: boost/pool/pool.hpp
 ===================================================================
---- boost/pool/pool.hpp	(revision 78317)
-+++ boost/pool/pool.hpp	(revision 78326)
+--- /boost/pool/pool.hpp	(revision 78317)
++++ /boost/pool/pool.hpp	(revision 78326)
 @@ -27,4 +27,6 @@
  #include <boost/pool/poolfwd.hpp>
  

boost-1.76.0-fix-narrowing-conversions-for-ppc.patch

@@ -13,8 +13,8 @@ Fixes #29
 
 diff --git a/include/boost/numeric/interval/detail/ppc_rounding_control.hpp b/include/boost/numeric/interval/detail/ppc_rounding_control.hpp
 index 87fe8ee..99f9986 100644
---- a/include/boost/numeric/interval/detail/ppc_rounding_control.hpp
-+++ b/include/boost/numeric/interval/detail/ppc_rounding_control.hpp
+--- include/boost/numeric/interval/detail/ppc_rounding_control.hpp
++++ include/boost/numeric/interval/detail/ppc_rounding_control.hpp
 @@ -28,10 +28,10 @@ typedef union {
    double dmode;
  } rounding_mode_struct;

boost-1.76.0-fix_multiprecision_issue_419-ppc64le.patch

@@ -8,10 +8,10 @@ Subject: [PATCH] Update gcc Intel intrinsic usage config. Fixes
  include/boost/multiprecision/cpp_int/intel_intrinsics.hpp | 6 +++++-
  1 file changed, 5 insertions(+), 1 deletion(-)
 
-diff --git a/include/boost/multiprecision/cpp_int/intel_intrinsics.hpp b/include/boost/multiprecision/cpp_int/intel_intrinsics.hpp
+diff --git a/boost/multiprecision/cpp_int/intel_intrinsics.hpp b/boost/multiprecision/cpp_int/intel_intrinsics.hpp
 index eb4624bb4..37717cd51 100644
---- a/include/boost/multiprecision/cpp_int/intel_intrinsics.hpp
-+++ b/include/boost/multiprecision/cpp_int/intel_intrinsics.hpp
+--- include/boost/multiprecision/cpp_int/intel_intrinsics.hpp
++++ include/boost/multiprecision/cpp_int/intel_intrinsics.hpp
 @@ -19,7 +19,11 @@
  // If this is GCC/clang, then check that the actual intrinsic exists:
  //

boost.spec

@@ -675,23 +675,9 @@ Historically, B2 was based on on FTJam and on Perforce Jam but has grown
 a number of significant features and is now developed independently.
 
 %prep
-%setup -q -n %{toplev_dirname}
+%autosetup -n %{toplev_dirname} -p1
 find ./boost -name '*.hpp' -perm /111 | xargs --no-run-if-empty chmod a-x
 
-%patch15 -p0
-%patch51 -p1
-%patch105 -p1
-%patch106 -p1
-%patch107 -p1
-%patch88 -p1
-%patch94 -p1
-%patch102 -p1
-%patch103 -p2
-%patch104 -p2
-%patch108 -p1
-%patch109 -p1
-%patch110 -p1
-
 %build
 %set_build_flags
 # Dump the versions being used into the build logs.
@@ -1296,6 +1282,11 @@ fi
 %{_mandir}/man1/b2.1*
 
 %changelog
+* Tue Sep 12 2023 Kefu Chai <tchaikov@fedoraproject.org> - 1.78.0-14
+- Apply patch for Boost.Accumulators bug
+  See https://github.com/boostorg/accumulators/pull/54
+- Change spec file to use autospec for applying patches
+
 * Fri Sep 01 2023 Jonathan Wakely <jwakely@redhat.com> - 1.78.0-13
 - Fix outdated CI configuration