perf: opt vector operations in tx query path

[songgaoye]

Closes: #XXX

Description

all 3 remove(0) operations replaced with into_iter().next(). It can avoids the front-shift penalty, O(n) -> O(1)

---

PR author checklist:

• Added changelog entry, using unclog.
• Added tests: integration (for Hermes) or unit/mock tests (for modules).
• Linked to GitHub issue.
• Updated code comments and documentation (e.g., docs/).
  • If guide has been updated, tag GitHub user mircea-c
• Tagged one reviewer who will be the one responsible for shepherding this PR.

Reviewer checklist:

• Reviewed Files changed in the GitHub PR explorer.
• Manually tested (in case integration/unit/mock tests are absent).

[KirilMihaylov]

Using swap_remove will reduce the readability overhead as it's equivalent to the provided solution.

P.S.:
Leaving this review as a member of the @nolus-protocol dev team.

[KirilMihaylov]

Suggested change

	let tx = response.txs.into_iter().next().expect("tx_search was constrained to a single result");
	let tx = response.txs.swap_remove(0);

[KirilMihaylov]

On second look, change is unnecessary in the first place because the assertion will fail if there is more than one element.

[KirilMihaylov]

Suggested change

	let tx = response.txs.into_iter().next().expect("tx_search was constrained to a single result");
	let tx = response.txs.swap_remove(0);

[KirilMihaylov]

Suggested change

	let (first_event, _, _) = tx_events.into_iter().next().expect("tx_events is known to contain at least one entry");
	let (first_event, _, _) = tx_events.swap_remove(0);

[songgaoye]

Using swap_remove will reduce the readability overhead as it's equivalent to the provided solution.

P.S.: Leaving this review as a member of the @nolus-protocol dev team.

I don't think so.
the current solution with into_iter().next() is even better because:

1.Same O(1) performance as swap_remove
2.Consumes the vector - makes it clear the rest of the data is discarded
3.More explicit intent - the code shows we only care about the first element
4.No unnecessary swap - swap_remove still does a swap operation even though we're discarding the vector

[KirilMihaylov]

At (1):

This does not preserve ordering of the remaining elements, but is O(1). If you need to preserve the element order, use remove instead.

Source: std::vec::Vec::swap_remove

At (2):

{ tx_events }.swap_remove(0)

At (3):
In my opinion it's just as explicit, and what's more, swap_remove is less verbose.

At (4):
Yes, I admit that it will do one swap, but that's a trade-off justified by being more concise in my eyes.