Fix: allow deleting only last tx and duplicates

[katspaugh]

What it solves

Resolves #3551

How this PR fixes it

Only the last transaction in the queue and transactions with a duplicate nonce are allowed to be deleted.

I used useRecommendedNonce() - 1 to calculate the last transaction nonce.

If a tx is in principle deletable but isn't the last one or a duplicate, we show a tooltip.

[github-actions]

Branch preview

✅ Deploy successful!

https://delete_last--walletweb.review-wallet-web.5afe.dev/home?safe=eth:0xA77DE01e157f9f57C7c4A326eeE9C4874D0598b6

[github-actions]

ESLint Summary View Full Report

Annotations are provided inline on the Files Changed tab. You can also see all annotations that were generated on the annotations page.

Type	Occurrences	Fixable
Errors	0	0
Warnings	0	0
Ignored	0	N/A

• Result: ✅ success
• Annotations: 0 total

---

^{Report generated by eslint-plus-action}

[github-actions]

📦 Next.js Bundle Analysis for safe-wallet-web

This analysis was generated by the Next.js Bundle Analysis action. 🤖

This PR introduced no changes to the JavaScript bundle! 🙌

[github-actions]

Coverage report

St.❔	Category	Percentage	Covered / Total
🟡	Statements	79.09%	11194/14154
🔴	Branches	58.52%	2649/4527
🟡	Functions	66.03%	1802/2729
🟢	Lines	80.37%	10087/12551

Test suite run success

1419 tests passing in 198 suites.

Report generated by 🧪jest coverage report action from e4cbfd3

[francovenica]

If there are 2 tx with the same nonce, but one of them is a signed Rejection, then I think we shouldn't allow the tx deletion of the "good transaction", since the user would be left with only the rejection, and we know users still don't know that they can create a new tx with an old nonce, so chances are they will end up executing that deletion to get rid of it.

Beyond that the ticket is fine.

[katspaugh]

the user would be left with only the rejection, ..., so chances are they will end up executing that deletion to get rid of it.

That's the purpose of a rejection tx though, isn't it?