test: Speedup fee estimation functional test with batching #35976

The fee estimation functional test is currently the slowest one by a good margin. It is a bit annoying, because it also increases the total runtime of the functional tests.

It seems like most of the slowness comes from the transactions propagating between the nodes. This patch helps them do that by submitting them directly to all the nodes. Also take this opportunity to batch the transaction submissions.

On my machine this speeds up the fee estimation functional test from around 71 seconds to 25 seconds.

<!--e57a25ab6845829454e8d69fc972939a-->

The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

<!--006a51241073e994b41acfe9ec718e94-->

Code Coverage & Benchmarks

For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/35976.

<!--021abf342d371248e50ceaed478a90ca-->

Reviews

See the guideline and AI policy for information on the review process.

If your review is incorrectly listed, please copy-paste <code>&lt;!--meta-tag:bot-skip--&gt;</code> into the comment that the bot should ignore.

<!--174a7506f384e20aa4161008e828411d-->

Conflicts

Reviewers, this pull request conflicts with the following ones:

• #34075 (fees: Introduce Mempool Based Fee Estimation to reduce overestimation by ismaelsadeeq)

If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

<!--5faf32d7da4f0f540f40219e4f7537a3-->

Concept ACK

tACK b3d77ea0275e56ad5aca442100b66789b5c08ec5

feature_fee_estimation.py passed on tip (72s) vs parent without the patch (92s). Slower absolute times than in the PR description, but still a clear improvement.

Batching to all nodes via create_self_transfer + node.batch matches the existing transact_and_mine pattern.

ACK b3d77ea0275e56ad5aca442100b66789b5c08ec5

The dropped confirmed_only=True is fine.

The old call was send_self_transfer(..., confirmed_only=True, utxo_to_spend=self.confutxo.pop(0)). confirmed_only is only consulted inside get_utxo(), which is short-circuited whenever utxo_to_spend= is passed explicitly:

# wallet.py
utxo_to_spend = utxo_to_spend or self.get_utxo(confirmed_only=confirmed_only)

Since utxo_to_spend is always supplied here, get_utxo() never runs and confirmed_only was already a no-op. Confirmedness is instead guaranteed by construction:

Benchmark result is almost identical to the OP master - 68.87 s
This PR 25.37 s

review ACK b3d77ea0275e56ad5aca442100b66789b5c08ec5 🐇

<details><summary>Show signature</summary>

Signature:

untrusted comment: signature from minisign secret key on empty file; verify via: minisign -Vm "${path_to_any_empty_file}" -P RWTRmVTMeKV5noAMqVlsMugDDCyyTSbA3Re5AkUrhvLVln0tSaFWglOw -x "${path_to_this_whole_four_line_signature_blob}"
RUTRmVTMeKV5npGrKx1nqXCw5zeVHdtdYURB/KlyA/LMFgpNCs+SkW9a8N95d+U4AP1RJMi+krxU1A3Yux4bpwZNLvVBKy0wLgM=
trusted comment: review ACK b3d77ea0275e56ad5aca442100b66789b5c08ec5 🐇
NwwFuvAZFroztCRi4rMYt4vLjWp8GxPOFa5UkTebzP0S33KRO+x8RntRuHUY9QLgOq+aPD2n0nqdfGKbuUpGBA==

</details>