Merge pull request 'Pass CMake arguments to mutation builds' (#30) from 10 into main

Reviewed-on: #30
Reviewed-by: andrew <andrew@aklabs.net>

CMakeLists.txt

@@ -1,6 +1,24 @@
 cmake_minimum_required(VERSION 3.10)
-# Version changelog notices DO NOT GO HERE. Put them in UPGRADING.md instead.
+# 1.0.0 replaced the consumer-sized __AKERR_ERROR_NAMES array with private
-project(akerror VERSION 2.0.3 LANGUAGES C)
+# storage. 2.0.0 makes the library thread safe, which is a second ABI break in
+# the same places: akerr_last_ignored became thread-local storage, and
+# ENSURE_ERROR_READY no longer takes the pool reference that akerr_next_error()
+# now takes for it. Consumer code compiled against a 1.x header would
+# double-count every reference. Hence the major bump and the SOVERSION, so a
+# stale installed libakerror.so cannot be silently paired with new headers.
+# 2.0.1 fixes the unhandled-error exit code, which reported success for any
+# status whose low byte was zero. It adds akerr_exit() but breaks nothing: the
+# soname is unchanged and no existing entry point changed shape.
+# 2.0.2 fixes the AKERR_USE_STDLIB=OFF build, which did not compile at all
+# (issue #12): untangles the header's includes, defines the AKERR_RUNTIME_HEADER
+# freestanding contract, retires PATH_MAX in favor of the AKERR_MAX_ERROR_FNAME_LENGTH
+# build option (default unchanged, so this is not an ABI break), and fails the
+# configure instead of the build when AKERR_THREADS would resolve to pthread
+# under AKERR_USE_STDLIB=OFF. ENSURE_ERROR_READY's pool-exhaustion path now
+# calls akerr_exit() instead of exit(1) directly, which changes that exit code
+# from 1 to AKERR_EXIT_STATUS_UNREPRESENTABLE (125) -- a deliberate behavior
+# change, not an ABI break: no soname move, no entry point changed shape.
+project(akerror VERSION 2.0.2 LANGUAGES C)
 
 include(GNUInstallDirs)
 include(CMakePackageConfigHelpers)