Close opt-out gaps between Drip and Constant Contact

**Objective:** List every person who opted out in one platform but can still be emailed by the other, looking back `{lookback-days}` days (default 30). Report only; change nothing.

**Step 1: Constant Contact opt-outs**

Call `constant-contact-get-campaign-unsubscribe` with `created-date.from` set to today minus the lookback. For each distinct `email-address`, call `drip-get-subscriber` with `email` set to that address. Keep the ones where Drip returns a person whose `status` is not unsubscribed. Record the Constant Contact `reason` and `created-date`.

**Step 2: Drip opt-outs**

Call `drip-get-subscriber` with `status` set to unsubscribed. Drip has no unsubscribe date filter, so use every result. For each `email`, call `constant-contact-get-contact` with `email-address` set to that address. Keep the ones where Constant Contact returns a contact whose `permission-to-send` is explicit or implicit.

**Step 3: Report**

Two tables:

1. **Unsubscribed in Constant Contact, still active in Drip:** email, Constant Contact unsubscribe date and reason, Drip `person-id`, Drip `status`.
2. **Unsubscribed in Drip, still mailable in Constant Contact:** email, Drip `person-id`, Constant Contact `contact-id`, `permission-to-send`.

Then one line each with the totals checked and the totals out of step. Suggest the `apply-constant-contact-opt-outs-to-drip` prompt for the first table.

**Rules & Guidelines:**

* `drip-get-subscriber` is on the drip server. `constant-contact-get-campaign-unsubscribe` and `constant-contact-get-contact` are on the constant-contact server.
* Match on email only, case-insensitively. Never match on name.
* This prompt is read-only. Do NOT call `drip-unsubscribe-subscriber`, `drip-update-subscriber` or `constant-contact-update-contact`, or any other action.

**Validation:**

* An address with no record in the other system is not out of step; count it as absent and leave it out of the tables.
* If a step returns nothing, say so for that direction and carry on with the other.
* If Step 2 returns more than 200 people, check the first 200, say how many were left unchecked, and suggest narrowing with a domain.