fix: propagate handler failures to CLI exit codes

[sadsfae]

Fixes #568

Summary

execute_badfish() in src/badfish/main.py invoked most handlers with a bare await badfish.<handler>(), discarding their return value. When a handler returned False to signal a failed operation, result stayed True and the CLI exited 0 (success) even though the command actually failed.

This change captures each operation-failure handler's return value and, on False, sets result = False, which drives main() to return exit code 1. Success (returning True, or None in a few handlers) leaves result = True -> exit 0.

Wrapped handlers (False == operation failed)

• check_schedule_job_status
• reset_idrac
• reset_bmc
• get_power_consumed_watts
• list_interfaces
• list_gpu
• mount_virtual_media
• unmount_virtual_media
• set_bios_password
• remove_bios_password
• get_scp_targets
• export_scp
• import_scp
• get_nic_fqdds
• get_nic_attribute / get_nic_attribute_info
• set_nic_attribute

Informational-False handlers intentionally left untouched

A False return does not mean a failure for these, so they must continue to exit 0:

• check_virtual_media — False = no virtual media mounted (not an error)
• check_remote_image — False = not attached / unsupported (informational)
• list_job_queue
• check_boot
• take_screenshot — already raises BadfishException on failure
• get_sriov_mode
• power-state queries (get_power_state)

Design note

Failures are propagated via the result = False channel — the same channel the already-fixed boot_to path uses — rather than raising BadfishException. Every wrapped handler already logs its own specific error before returning False; raising an exception through the dispatch would emit a duplicate error line and break existing CLI-output assertions. result = False produces a single clean error message and still yields exit code 1.

Tests

Added tests/test_exit_code_propagation.py (35 cases) proving:

• handler returns False -> execute_badfish returns ("host", False) -> CLI exit 1
• handler returns True -> exit 0
• handlers returning None on success (e.g. set_bios_password) are not misreported as failures
• check_virtual_media / check_remote_image returning False are not treated as failures

Verified:

• PYTHONPATH=src python -m pytest tests/test_exit_code_propagation.py -v -> 35 passed
• PYTHONPATH=src python -m pytest tests/ -q -> 491 passed, 1 pre-existing unrelated failure (tests/test_progress.py::test_terminal_renders_progress_when_enabled)
• The new failure-propagation tests fail on the pre-fix code (regression-valid).

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Flag	Coverage Δ
unittests	94.90% <100.00%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/badfish/main.py	93.38% <100.00%> (+0.04%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.