rpc: Avoid getchaintxstats invalid results

[maflcko]

The getchaintxstats RPC reply during AU background download may return non-zero, but invalid, values for window_tx_count and txrate.

For example, txcount may be zero for a to-be-downloaded block, but may be non-zero for an ancestor block which is already downloaded. Thus, the values returned may be negative (and cause intermediate integer sanitizer violations).

Also, txcount may be accurate for the snapshot base block, or a descendant of it. However it may be zero for an ancestor block that still needs to be downloaded. Thus, the values returned may be positive, but wrong.

Fix all issues by skipping the returned value if either txcount is unset (equal to zero).
Also, skip txcount in the returned value, if it is unset (equal to zero).

Fixes #29328

[DrahtBot]

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

Code Coverage

For detailed information about the code coverage, see the test coverage report.

Reviews

See the guideline for information on the review process.

Type	Reviewers
ACK	fjahr

If your review is incorrectly listed, please react with 👎 to this comment and the bot will ignore it on the next update.

Conflicts

Reviewers, this pull request conflicts with the following ones:

• #29656 (chainparams: Change nChainTx type to uint64_t by fjahr)

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.

[fjahr]

Concept ACK, could we add a test for this?

[maflcko]

A test is already present in test/functional/feature_assumeutxo.py, see the diff. You can also test it by running this pull against a bitcoind compiled from master and observing the sanitizer crash and the test failure.

[fjahr]

tACK fa28613

I have written some additional coverage to test the behavior and build my understanding, it could be added here or I can make a follow-up: fjahr@03a0f31

[maflcko]

03a0f31

I think assert_equal can generally not be used on floating point values, no?

[fjahr]

re-ACK fa73710

I think assert_equal can generally not be used on floating point values, no?

Here is a version with assert_approx: fjahr@24e9a37

[maflcko]

@fjahr Thanks! I've pushed your test commit with small fixups.

[DrahtBot]

🚧 At least one of the CI tasks failed. Make sure to run all tests locally, according to the
documentation.

Possibly this is due to a silent merge conflict (the changes in this pull request being
incompatible with the current code in the target branch). If so, make sure to rebase on the latest
commit of the target branch.

Leave a comment here, if you need help tracking down a confusing failure.

_{Debug: https://github.com/bitcoin/bitcoin/runs/25183174149}

[maflcko]

The CI failure can be ignored during review. It is fixed in #30144

[fjahr]

re-ACK ef692e5

Also reviewed #30144 to confirm the failure is ok.