petri: allow test to override default VTL2 allocation scheme

[mattkur]

Promote the settings that change how VTL2 memory is allocated to the Petri VM objects, rather than being hidden in the depths of the OpenVMM config. This is a convenience for additional tests coming as part of #2215 .

[Copilot]

Pull Request Overview

This PR promotes VTL2 memory allocation configuration settings from internal OpenVMM config to the Petri VM builder API, making it easier to test different VTL2 memory allocation schemes. The changes extract hardcoded allocation logic and expose it through a builder method.

Key changes:

• Added with_vtl2_base_address_type() builder method to configure VTL2 memory allocation
• Refactored existing tests to use the new API instead of direct config modification
• Moved default VTL2 base address type logic from construct code to the Firmware initialization

Reviewed Changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
petri/src/vm/mod.rs	Adds vtl2_base_address_type field to OpenHclConfig and with_vtl2_base_address_type() builder method; moves default allocation logic to Firmware::openhcl_uefi()
petri/src/vm/openvmm/construct.rs	Removes hardcoded VTL2 allocation logic and uses config from OpenHclConfig; adds logic to disable late map based on allocation type
vmm_tests/vmm_tests/tests/tests/x86_64/openhcl_linux_direct.rs	Refactors test to use new builder API instead of modifying backend config directly
vmm_tests/vmm_tests/tests/tests/x86_64/openhcl_uefi.rs	Adds VTL2 allocation configuration to nvme_relay test
petri/src/vm/hyperv/mod.rs	Adds TODO comment for future Hyper-V backend support

[github-actions]

At least one Petri test failed.

[smalis-msft]

I think this looks good, but i'd like a sanity check from @chris-oo too.