[ORCA] Fix flaky "Invalid key is inaccessible" fallback (#15147)

[fanfuxiaoran]

In CI pipeline there were occassional test failures due to ORCA fallback with following stacktrace.

```
+INFO:  GPORCA failed to produce a plan, falling back to planner
+DETAIL:  CSyncHashtable.h:109: Failed assertion: IsValid(key) && "Invalid key is inaccessible"
+Stack trace:
+1    gpos::CException::Raise + 227
+2    <symbol not found> + 15235666
+3    gpos::CMemoryPoolManager::CreateMemoryPool + 653
+4    gpos::CAutoMemoryPool::CAutoMemoryPool + 34
+5    gpopt::CColumnFactory::CColumnFactory + 80
+6    gpopt::COptCtxt::PoctxtCreate + 77
+7    gpopt::CAutoOptCtxt::CAutoOptCtxt + 54
+8    gpopt::COptimizer::PdxlnOptimize + 411
+9    COptTasks::OptimizeTask + 850
+10   gpos::CTask::Execute + 52
+11   gpos::CWorker::Execute + 36
+12   gpos::CAutoTaskProxy::Execute + 97
+13   gpos_exec + 557
```

Core dump of failure showed CMemoryPool::m_hash_key had invalid key value 0xffffffff. Hence, the query raised an assertion error and fell back to PLANNER.

Issue is that CMemoryPool::m_hash_key was never directly initialized. This suggests that it was using uninitialized memory to produce randomness in the key. When that memory contains 0xffffffff in just the right place, then the value of the CMemoryPool::m_hash_key is an invalid key and ORCA falls back.

Following is patch that demonstrates the issue:
```
diff src/backend/utils/mmgr/aset.c
@@ -989,6 +989,8 @@ AllocSetAlloc(MemoryContext context, Size size)

                MEMORY_ACCOUNT_INC_ALLOCATED(set, chunk->size);

+               memset((char *) AllocChunkGetPointer(chunk), 0xFFFFFFFF, size);
+
                return AllocChunkGetPointer(chunk);
        }
```

A few lines above that patch, you can see that when compiled with RANDOMIZE_ALLOCATED_MEMORY the memory is randomly initialied. So we can make no assumptions about the uninitialied memory; meaning that 0xffffff is valid.

Note: Seemed this failure manifested more commonly with JIT ICW runs. (cherry picked from commit 2c7152f46aced9328d86dc1025d0395fcf467455)

fix #ISSUE_Number

---

Change logs

Describe your change clearly, including what problem is being solved or what feature is being added.

If it has some breaking backward or forward compatibility, please clary.

Why are the changes needed?

Describe why the changes are necessary.

Does this PR introduce any user-facing change?

If yes, please clarify the previous behavior and the change this PR proposes.

How was this patch tested?

Please detail how the changes were tested, including manual tests and any relevant unit or integration tests.

Contributor's Checklist

Here are some reminders and checklists before/when submitting your pull request, please check them:

• Make sure your Pull Request has a clear title and commit message. You can take git-commit template as a reference.
• Sign the Contributor License Agreement as prompted for your first-time contribution(One-time setup).
• Learn the coding contribution guide, including our code conventions, workflow and more.
• List your communication in the GitHub Issues or Discussions (if has or needed).
• Document changes.
• Add tests for the change
• Pass make installcheck
• Pass make -C src/test installcheck-cbdb-parallel
• Feel free to request cloudberrydb/dev team for review and approval when your PR is ready🥳

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
<sub>You have signed the CLA already but the status is still pending? Let us recheck it.</sub>

[my-ship-it]

Seems icw-parallel-test still fail, so need to fix

[fanfuxiaoran]

Seems icw-parallel-test still fail, so need to fix

Done