Adjust boolean switch aliases to follow logical defaults

[t-kalinowski]

Summary

This makes boolean switch help default-driven without making parsing brittle. Help now shows the most useful spelling for changing the default, while the parser accepts common explicit boolean forms users are likely to try.

User-facing changes

• FALSE, TRUE, and NA logical defaults are boolean switches by default.
• Help shows --foo for foo <- FALSE, --no-foo for foo <- TRUE, and --foo / --no-foo for foo <- NA.
• Switch help no longer generates default text such as [default: true], [enabled by default], or toggle hints. It shows the flag spelling plus any user-provided description.
• Boolean switches accept bare toggles and explicit values: --foo, --no-foo, --foo=true, --foo=false, --foo true, and --foo false where the alias is enabled.
• Bare --foo and --no-foo count as supplied values; omitted foo <- NA remains NA.
• Help display is intentionally narrower than parser acceptance. For example, foo <- TRUE normally shows --no-foo, but --foo=false is still accepted as an explicit value for foo.
• #| negative_alias: false only disables the generated --no-* spelling. It does not disable --foo or explicit value forms such as --foo=false.
• If foo <- TRUE uses #| negative_alias: false, help falls back to --foo <FOO> because there is no default-changing bare alias left.
• #| arg_type: option remains available when a boolean input should require an explicit value; generated --no-* spellings are rejected for these explicit bool options.
• The README links to a new boolean behavior table in docs/boolean-options.md.

Internal changes

• Split boolean switch handling into accepts_* helpers for parser acceptance and shows_* helpers for help display.
• Parse optional boolean values after positive switches while preserving positional arguments through pushback.
• Reject disabled negative aliases and explicit bool-option --no-* spellings before positional fallback.
• Keep short aliases aligned with the default-driven spelling shown in help.
• Add regression coverage for tri-state switches, disabled negative aliases, explicit boolean values, required boolean option values, short aliases, positional fallback rejection, and switch help output.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 9398a7d5d3

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: cbc5784f54

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 16dd0bf969

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".