[thebrandonlucas]

Problem

voltage environments list --org ORG --all failed with exit code 2: unexpected argument '--all' found. #94 made API lists without paging, such as wallets list, accept --all and ignore it. The local lists are clap-derived subcommands, not registry operations, so #94 did not reach them. organizations list and profiles list had the same failure.

Change

Command	Before	After
environments list --all	exit 2, unexpected argument	accepted, ignored
organizations list --all	exit 2, unexpected argument	accepted, ignored
profiles list --all	exit 2, unexpected argument	accepted, ignored
profiles get NAME --all, context --all	rejected	still rejected

• src/cli.rs: the --all argument from feat(cli): accept --all on lists that return every item at once #94 moves into all_at_once_arg(). The registry lists and the two local List variants (ListCommand::List, ProfileCommand::List) use it. The help text is now "No effect: this list shows every item at once", which is also true for the local profiles list.
• Tests: all_is_accepted_by_every_list_command walks the whole command tree and fails if any list subcommand has no --all, so a new list cannot miss it. all_is_ignored_by_local_lists_and_rejected_by_other_local_commands tests the three local lists and the rejection path for local commands that are not lists. Both tests fail without the fix.
• CHANGELOG.md (Unreleased) and docs/guides/requests.md (Paging) are updated.

About --wallet <UUID> in the usage line

That usage line is not a real requirement. When clap suggests a similar flag (--all → --wallet), Parser::did_you_mean_error in clap_builder 4.6.7 adds the suggested argument to the matches "to build a proper usage string". So the usage line names the suggestion as if it were given. voltage environments list --walle shows the same line. Now that --all is accepted, this command no longer prints that suggestion. The clap behavior for other typos is out of scope here.

Closes #119
Linear: DEV-2954

Checks

• cargo fmt --check, cargo clippy --locked --all-targets -- -D warnings, cargo machete --with-metadata, cargo test --locked (199 unit + 149 integration tests pass), python3 scripts/check-coverage.py: all pass.
• I removed the fix and confirmed that both new tests fail.
• Manual run: environments list --all and organizations list --all now get past parsing and stop at the missing credential (exit 3). profiles get x --all is still rejected.
• CRAP report (scripts/crap-report.sh, cargo-llvm-cov 0.8.5, knots 1.16.0): 0 violations. The highest score is 18.10 (src/api.rs wait_for_payment, unchanged). No function scores 20 or more. endpoint in src/cli.rs is at 18.00 and this PR does not change it.
• The voltage-pr-review self-review led to one fix: the help text no longer says "response", because profiles list is local.
• Not run: the Windows CI job (no Windows host). There are no live or staging checks because the change is offline only.

[tee8z]

Looks good. One all_at_once_arg() serves the registry lists and the local lists. The tree walk in all_is_accepted_by_every_list_command keeps a future list from missing --all, and the second test covers the rejection path. The changelog and the Paging docs are updated.