fuzz: track chanmon claims on PaymentClaimed events

[joostjager]

Only record receiver claims in the fuzz payment tracker after the node emits PaymentClaimed. This keeps the tracker aligned with durable claim completion instead of a claim_funds call that a restart may roll back.

[ldk-reviews-bot]

👋 I see @wpaulino was un-assigned.
If you'd like another reviewer assignment, please click here.

[ldk-claude-review-bot]

I've thoroughly reviewed the full diff and the surrounding context. The change is small and correct:

1. Moving claimed_payment_hashes.insert(payment_hash) from the claim_payment method (at claim_funds call time) to a new mark_claimed method called from the PaymentClaimed event handler aligns the tracker with durable state.
2. The mark_claimed method is idempotent (HashSet insert).
3. The rollback logic at line 2953 (claimed_payment_hashes.remove) still correctly handles the case where a sender node is reloaded and its sent payments are rolled back.
4. The receiver-side reload case is now actually handled more correctly — if a receiver node is reloaded before PaymentClaimed fires, the hash was never added to claimed_payment_hashes, so there's nothing stale to clean up.

No issues found.

[ldk-reviews-bot]

🔔 1st Reminder

Hey @TheBlueMatt! This PR has been waiting for your review.
Please take a look when you have a chance. If you're unable to review, please let us know so we can find another reviewer.

[ldk-reviews-bot]

🔔 2nd Reminder

Hey @TheBlueMatt! This PR has been waiting for your review.
Please take a look when you have a chance. If you're unable to review, please let us know so we can find another reviewer.

[TheBlueMatt]

I mean yes, I believe this should be required, but I'm more interested in why we didn't run into failures without this? Any chance we could debug that and figure out what coverage we're missing?

[ldk-reviews-bot]

👋 The first review has been submitted!

Do you think this PR is ready for a second reviewer? If so, click here to assign a second reviewer.

[joostjager]

I mean yes, I believe this should be required, but I'm more interested in why we didn't run into failures without this? Any chance we could debug that and figure out what coverage we're missing?

I did try that: #4631 (comment)

It seems the reload replay prevented failures from happening.