fix: make drag coordinate order explicit

[nightcityblade]

What & why

Change Terminal::drag to accept four explicitly ordered column/row arguments, matching the other mouse APIs. This prevents the row-first tuple returned by Screen::find from being passed directly and silently transposed. The relevant docs and drag tests now use the unambiguous spelling.

Closes #257.

Checklist

• Linked an issue (or explained above why none exists)
• Tests added/updated for the change
• cargo fmt --all and cargo clippy --workspace --all-targets --all-features are clean
• All commits are signed off (git commit -s) — see CONTRIBUTING.md §5
• No AI attribution trailers (no AI co-authors, "Generated with" footers, or bot identities) — see CONTRIBUTING.md §6
• CHANGELOG.md updated under [Unreleased] (user-facing changes only)
• Snapshot changes (if any) were reviewed with cargo insta review, not blind-accepted (N/A: no snapshots changed)

[vyncint]

Reviewed and verified. The signature change does exactly what the issue asked.

What I checked beyond CI:

• The transposition is now a compile error rather than a silent wrong gesture: passing a Screen::find result straight in fails with "this method takes 5 arguments but 3 arguments were supplied".
• The drag suite passes unchanged, including the per-cell motion test and the off-grid endpoint test, so the four arguments reach check_mouse_in_grid in the same order the tuples did.
• Reconstructing from/to inside the body keeps the diff to the boundary, which is the right call.

Taking option 1 from the issue was the better of the two, and the cross-reference you added on Screen::find is the half I would have forgotten.

One more change, alongside the shared one below

skills/termlens/SKILL.md landed on main a few hours after you opened this, and it documents the old two-tuple signature in three places. It is the file coding agents copy from, so it has to move with the API:

• line 99, in the coordinate-order rule: `drag(button, (col, row), (col, row))` becomes `drag(button, from_col, from_row, to_col, to_row)`, and the sentence after it that says "the tuple types match and the axes do not" now needs to say that the four arguments make the transposition unwritable.
• line 369, in the API cheat sheet: `drag(MouseButton::Left, (c, r), (c, r))`.
• line 414, in the pitfalls table: the row whose left column is `t.drag(b, s.find("x").unwrap(), …)`.

.github/scripts/check-skill-snippets.sh compiles the skill's ```rust blocks, but these three are prose and tables, so CI cannot catch them. Sorry for the collision — it is entirely an accident of merge order, not something you could have seen when you opened this.

---

One thing to fix, and it is the same in all five of your PRs

The CHANGELOG entry landed inside the published ## [0.9.0] - 2026-09-05 section rather than under ## [Unreleased]. Easy to miss: [Unreleased] was empty when you branched, so the nearest heading with content under it belongs to 0.9.0.

It matters for two reasons. 0.9.0 is already on crates.io and its GitHub Release notes were generated from that section, so the entry now describes a release that does not contain the change. And docs/RELEASING.md cuts the next release by moving [Unreleased] down into a new version heading — an entry parked in [0.9.0] would never appear in any release notes at all.

Move it up under ## [Unreleased], adding the ### Added / ### Changed / ### Fixed sub-heading there if it does not exist yet. main has moved on by a commit since you branched, so a rebase is wanted anyway:

git fetch origin && git rebase origin/main

Thanks for these — five well-scoped fixes with tests and docs, and the PR bodies say what you decided and why. That is exactly the shape this repository asks for.

[nightcityblade]

Addressed both follow-ups in f8c78e4: rebased and moved the changelog entry under Unreleased → Changed, and updated all three skill references to the four-argument drag API.

Validation: cargo fmt --check, cargo test --workspace --all-features, and .github/scripts/check-skill-snippets.sh pass.

[vyncint]

Thanks — both points addressed, and the skill update is thorough. You caught all three places, and the rewrites are accurate rather than mechanical: the rule now explains what the four arguments buy, and the pitfalls row describes the mistake in words instead of showing a call that no longer compiles.

Re-verified on this head: passing a Screen::find result straight into drag is a compile error, the drag suite passes unchanged including the per-cell motion and off-grid endpoint cases, and the skill's compile guard is green.

I rebased the branch onto main and resolved the changelog collision with #267 and #268, which landed ahead of it. Your commit, author and sign-off are untouched. Merging.