Fix router fuzz failure due to LengthLimitedReader

[valentinewallace]

We recently switched the decode_msg macro in the router fuzz target from reading from a Cursor to reading from a slice. This caused a failure because the slice advances its pointer as it is being read from, so asserting that the length of the slice is equal to the length of the message that was read no longer works. Instead assert that the original fuzz data length is equal to the length of the message that was read.

Closes #3692

[ldk-reviews-bot]

👋 Thanks for assigning @joostjager as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

[TheBlueMatt]

It looks like the intention behind the assert was to check that all of data was consumed (i.e. we should assert!(reader.is_empty()), but it was previously broken. The old and suggested assert here are basically useless.

[joostjager]

Re-request review after Matt's point has been addressed.

[valentinewallace]

It looks like the intention behind the assert was to check that all of data was consumed (i.e. we should assert!(reader.is_empty()), but it was previously broken. The old and suggested assert here are basically useless.

🤦‍♀️ Thanks!

[joostjager]

Assuming the fuzzer is not failing anymore...

[TheBlueMatt]

Same, we can fix it more if its still broken.