Add ASAN support and fix memleaks (main)

[alanking]

Addresses #286
Addresses #287
Addresses #288

Cherry-pick of #289

Tests are not passing due to #290. However, there isn't much difference between this and 4-3-stable, so maybe it's okay...

diff --git a/CMakeLists.txt b/CMakeLists.txt
index b2d4a79..9197f7d 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,6 +1,6 @@
 cmake_minimum_required(VERSION 3.5.0 FATAL_ERROR) #CPACK_DEBIAN_<COMPONENT>_PACKAGE_NAME
 
-set(IRODS_MINIMUM_VERSION "4.3.3")
+set(IRODS_MINIMUM_VERSION "4.90.0")
 find_package(IRODS "${IRODS_MINIMUM_VERSION}" REQUIRED)
 set(IRODS_PLUGIN_REVISION "0")
 set(IRODS_PLUGIN_VERSION "${IRODS_MINIMUM_VERSION}.${IRODS_PLUGIN_REVISION}")
@@ -22,7 +22,6 @@ project(irods_capability-storage_tiering
 include("${IRODS_TARGETS_PATH}")
 
 include(GNUInstallDirs)
-include(UseLibCXX)
 
 if (NOT CMAKE_CONFIGURATION_TYPES AND NOT CMAKE_BUILD_TYPE)
        set(CMAKE_BUILD_TYPE Release CACHE STRING "Choose the type of build {Debug, Release}." FORCE)
diff --git a/irods_consortium_continuous_integration_build_hook.py b/irods_consortium_continuous_integration_build_hook.py
index 73bf2bf..b94bf7d 100644
--- a/irods_consortium_continuous_integration_build_hook.py
+++ b/irods_consortium_continuous_integration_build_hook.py
@@ -12,11 +12,10 @@ def add_cmake_to_front_of_path():
 
 def install_building_dependencies(externals_directory):
     externals_list = [
-        'irods-externals-boost-libcxx1.81.0-1',
-        'irods-externals-clang-runtime13.0.1-0',
+        'irods-externals-boost1.81.0-1',
         'irods-externals-clang13.0.1-0',
         'irods-externals-cmake3.21.4-0',
-        'irods-externals-fmt-libcxx8.1.1-1',
+        'irods-externals-fmt8.1.1-1',
         'irods-externals-json3.10.4-0'
     ]
     if externals_directory == 'None' or externals_directory is None:

[trel]

Yep - looks good.

Does it make sense to see if #290 can be addressed along with this? Then they can all go in together, passing?

[korydraughn]

Code changes look good.

Deferring to @alanking on the question about #290.

[alanking]

Does it make sense to see if #290 can be addressed along with this? Then they can all go in together, passing?

Either way works for me. The tests are currently failing with or without this PR, so it doesn't make much difference. This is just a cherry-pick so that the changes are reflected in all the relevant branches.

[korydraughn]

Ok. Let's get this PR in and handle the tests in a separate PR. That reduces the number of changes for the reviewer to track.

[korydraughn]

Is this PR complete?

[alanking]

Yes, I think so. Changes look the same

[korydraughn]

Pound it if you're happy with it.

[alanking]

#'d. Mergin