fix: Verify MemoryShm::byte_size inside shared memory boundary

[yinggeh]

What does the PR do?

When attacker registers the same shm created by python backend, they can overwrite MemoryShm::byte_size data with a very large value. Identity model will read a large chunk of sensitive data (e.g. glibc.so) as input tensor, copy to the output tensor and send back to the client.

Checklist

• PR title reflects the change and is of format <commit_type>: <Title>
• Changes are described in the pull request.
• Related issues are referenced.
• Populated github labels field
• Added test plan and verified test passes.
• Verified that the PR passes existing CI.
• Verified copyright is correct on all changed files.
• Added succinct git squash message before merging ref.
• All template sections are filled out.
• Optional: Additional screenshots for behavior/output changes with before/after.

Commit Type:

• fix

Related PRs:

Where should the reviewer start?

Test plan:

• CI Pipeline ID:
  31182731

Caveats:

Background

Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)

• closes GitHub issue: #xxx

[yinggeh]

No unit test because triton-inference-server/server#8273 makes the exploitation impossible.

[pskiran1]

LGTM

[yinggeh]

@Tabrizian @tanmayv25 @pskiran1 I pushed a new commit 595a488 to fix the CI. Somehow the setting of this repo does not enforce re-approval. Please add an emoji if the change looks good. Thanks.