[thebrandonlucas]

-o is now short for --output, as in voltage wallets list -o json.

In the CLI benchmark, agents tried -o json and got unexpected argument '-o'. Closes #85.

Stack: 1 of 6, based on master. The PRs above it don't depend on this change; they share the stack only to keep CHANGELOG.md ordered.

• User-visible change: CHANGELOG.md (Unreleased) is updated, and docs/reference/global-flags.md is regenerated with VOLTAGE_BLESS=1.
• The verification gate passes: fmt, clippy, machete, cargo test (252 tests) and check-coverage.py.
• Reviewed against master. A new unit test parses -o json both before and after the subcommand. The change adds no error or early-return path.
• CRAP on 79b7856: 0 functions at or above 20. The highest is wait_for_payment (src/api.rs) at 19.60.
• Checks not run: none. I confirmed that no other flag uses -o as its short form.

[tee8z]

Approved 79b7856 against master 520c775. -o is the only short form added, and clap's command-tree assertions run in the test suite, so a clash with another flag would fail there. The new test covers -o before and after the subcommand, and global-flags.md is regenerated.

Validation on this head, in nix develop: cargo fmt --check, cargo clippy --locked --all-targets -- -D warnings, cargo machete --with-metadata and python3 scripts/check-coverage.py passed. In cargo test --locked, all 132 unit tests passed. One integration test failed once: login_on_a_terminal_reports_why_no_profile_was_saved_and_still_succeeds timed out waiting for the command to exit. It comes from #75, and this PR does not touch it. It passed in the CRAP run's full suite on this head, and in 2 of 3 isolated reruns, with the failure at a different line each time. That is a timing flake under load, not this change. CRAP: 608 measured functions, none at or above 20, highest wait_for_payment at 19.60; 174 functions were unmeasured. CI is green.