AMDGPU: reduce hotswap helper duplication

[harsh-amd]

Summary:

• Share the hotswap uint64 overflow-checked add helper between ELF growth and entry-trampoline code.
• Collapse duplicated retargetCodeObject output-buffer allocation/copy handling.
• Keep behavior unchanged and independent of the hotswap lit-test PR.

This addresses the implementation-redundancy feedback from #3008 in a separate refactor PR. It is based directly on amd-staging and touches only hotswap implementation helper code, so it can be reviewed in parallel with #3098.

Verification:

• cmake --build build --target hotswap-rewrite -- -j8
• cmake --build build --target HotswapMCTests -- -j8
• build/bin/HotswapMCTests
• build/bin/llvm-lit -sv build/tools/comgr/test-lit --filter hotswap

[chinmaydd]

Thanks !

Also, request to keep the commit message to a minimal description of the change being made. A lot of the comments from the PR description can be elided.