[Nexthop][fboss2-dev] Add fboss2-dev config arp proactive subcommand

[vybhav-nexthop]

Pre-submission checklist

• I've ran the linters locally and fixed lint errors related to the files I modified in this PR. You can install the linters by running pip install -r requirements-dev.txt && pre-commit install
• pre-commit run

Summary

• Adds fboss2-dev config arp proactive [enabled|disabled] to the existing config arp command family
• Mutates SwitchConfig.proactiveArp (declared in switch_config.thrift) and persists via HITLESS config save — no agent restart required
• proactiveArp is not yet consumed by the agent (ApplyThriftConfig.cpp has no read, switch_state.thrift has no corresponding field); the CLI is wired ahead of agent support. A code comment flags this explicitly.

Changed files

File	Change
CmdConfigArp.cpp	Add proactive to valid attrs; parse enabled/disabled; dispatch to swConfig.proactiveArp()
CmdConfigArp.h	Add getBoolValue() / boolValue_; update addCliArg() help text to mention proactive
test/config/CmdConfigArpTest.cpp	Unit tests for arg validation + queryClient round-trip for proactive
test/integration_test/ConfigArpTest.cpp	Integration test: set/commit/verify round-trip via running agent config
test/integration_test/Fboss2IntegrationTest.h	Add two-arg getSwConfigField(field, default) overload + private getSwConfigFieldOpt<T> helper for optional/defaulted config fields

Fboss2IntegrationTest.h refactor

icmpV4UnavailableSrcAddress is optional in the agent config — devices that have never configured it have no entry in the JSON. The integration test needs to read the field with a fallback to the agent's implicit default (192.0.0.8) when absent.

To support this without duplicating the type-dispatch logic (bool/int/string/double), the existing getSwConfigField is refactored:

• A private getSwConfigFieldOpt<T> helper does the actual field lookup and type conversion, returning std::optional<T>.
• The original single-arg getSwConfigField delegates to it and throws when nullopt (required fields — existing callers unaffected).
• A new two-arg overload delegates to it and returns the caller-supplied default when nullopt (optional fields).

Integration Test Results — NH-4010-F

Ran ConfigArpTest suite on DUT after latest changes:

[==========] Running 5 tests from 1 test suite.
[  PASSED  ] ConfigArpTest.SetTimeout (159 ms)
[  PASSED  ] ConfigArpTest.SetAgeInterval (117 ms)
[  PASSED  ] ConfigArpTest.SetMaxProbes (118 ms)
[  PASSED  ] ConfigArpTest.SetStaleInterval (119 ms)
[  PASSED  ] ConfigArpTest.SetProactive (118 ms)
[==========] 5 tests from 1 test suite ran. (633 ms total)
[  PASSED  ] 5 tests.

Test plan

• Unit test: bazel test //fboss/cli/fboss2/test/config:CmdConfigArpTest
• Integration test on DUT: fboss2-dev config arp proactive enabled → commit → verify proactiveArp: true in running config → restore

Notes

proactiveArp is not yet consumed by the agent; the integration test only
proves the value round-trips through the config (set → commit → read back),
not that proactive ARP behavior changes.

Review Findings

Pre-publication review (3 correctness angles × 7 candidates, 1-vote verify).
Issues found and fixed before push:

• result.empty() fallback in queryClient replaced with explicit result assignment per branch (eliminates latent getValue() trap for future boolean attrs)
• Fboss2IntegrationTest.h: single-arg getSwConfigField restored to inline form with distinct "missing 'sw' object" vs "missing field" error messages
• Added EXPECT_THROW unit tests for getValue()/getBoolValue() cross-kind accessor guards

No issues above confidence threshold remain.

[joseph5wu]

@vybhav-nexthop Let's abandon this PR since this proactiveArp is never really used in agent. Your change here is basically no-op to the system and it could be confusing to exposing some cli which is not actually doing anything.

cc: @benoit-nexthop