fix(workspace): treat '/' and '\' as the same separator in root matching (#1718)

[umi008]

What does this PR do?

--approve-sensitive never lifted the sensitive-root refusal on Windows for paths under Program Files, because the grant is recorded from cbm_canonical_path (backslashes) while the indexer hands the candidate already normalized to /. ws_is_ancestor_or_equal compared the raw bytes, so the recorded grant never matched the candidate and the escape hatch stayed dead exactly where it is needed most.

The comparison now treats / and \ as the same separator on the component boundary, consistent with ws_is_sep everywhere else in this module. The containment rule (a separator byte is never part of a component name) is unchanged.

The ws_paths_equal home-directory rule in cbm_workspace_classify_root uses the same helper, so home detection becomes robust to separator spelling as well. No classification semantics changed beyond separator equivalence.

Checklist

• Every commit is signed off (git commit -s) — required, CI rejects unsigned commits (DCO, see CONTRIBUTING.md)
• Tests pass locally (make -f Makefile.cbm test)
• Lint passes (make -f Makefile.cbm lint-ci)
• New behavior is covered by a test (reproduce-first for bug fixes)

[github-actions]

Thanks for opening this — it has been seen, and it is queued.

This note is automated, but it is not a brush-off: it exists so you know where your PR stands instead of having to guess from silence.

Current review status: working through a backlog. 0.9.1-rc.1 is out, so the release freeze that held reviews is over — but it left a large queue of open pull requests behind it, and we are reading through them oldest-first. The background is in discussion #1144.

What that means for this PR, concretely:

• It will not be closed for inactivity. No stale bot touches pull requests here.
• It may still sit a while before a human reads it. That is on us, not on you.
• Older PRs are read first, so a recent one is not being skipped — it is behind a queue.

Things that will genuinely speed it up whenever review does happen:

• Keep it rebased on main — the tree is moving quickly right now, and a conflicting branch cannot be reviewed as the diff you intended.
• Get CI green, or say which failures you believe are pre-existing.
• Keep the change to one claim. Bundled features and refactors get split before they get merged, which costs you a round trip.
• Every commit needs a sign-off (git commit -s) — CI enforces DCO.

If this fixes a bug, a reproduction we can run is worth more than a description of the symptom.

Thanks for contributing, and sorry in advance for the wait.

[DeusData]

Thank you for finding this separator-normalization hole in the sensitive-root grant path. I checked current main: ws_is_ancestor_or_equal() trims and checks separators at boundaries, but compares the path prefix byte-for-byte, so a recorded backslash path cannot match the same Windows path normalized with forward slashes. That grounds both the approval and home-path mismatch described here.

I have labeled this as a high-priority Windows security bug and routed it for review. The Windows test shard is currently red, so that failure must be attributed before the PR can advance. Our review queue is full, so detailed review may take a little time. Thank you for keeping the containment boundary intact while narrowing the change to separator equivalence.