BIP: The Cat - Non-Monetary UTXO Cleanup

[ostromcode]

This PR proposes a new consensus (soft fork) BIP called "The Cat: Non-Monetary UTXO Cleanup".

The proposal:

• Defines NMUSet_snap as a snapshot-based set of Non-Monetary UTXOs (NMUs), derived solely from fully deterministic Ord 0.24.0 and Stamps-style classification rules applied to on-chain data, while nodes only need the resulting NMU_DATA blob and its hash to enforce the rule.
• Makes those NMUs permanently unspendable by consensus and permits their removal from the UTXO set.
• Uses a Binary Fuse Filter plus a false-positive exclusion list (NMU_DATA) to encode NMUSet_snap without requiring nodes to reindex.
• Targets dust-sized inscription/stamp UTXOs in a fixed height window, with a hard value threshold (< 1,000 sats), leaving ordinary monetary UTXOs untouched.

The draft is written in Markdown as bip-the-cat.md.
The Discussions-To: header points to the repository where the draft was initially developed and where additional charts and analysis live:
https://github.com/ostromcode/The-Cat

I attempted to submit an earlier version of this text to the Bitcoin-dev mailing list but the messages did not appear on the list. Until that is resolved, this PR and the linked repository are the primary venues for public review.

I welcome feedback in particular on:

• The economic/incentive framing of permanently disabling dust NMUs as a way to remove demand for inscription-style use of the UTXO set.
• The snapshot design and NMU_DATA construction (height window, Ord/Stamps pinning, Binary Fuse Filter + exclusion list), including any better alternatives for encoding NMUSet_snap.
• Consensus and deployment risks: edge cases I may have missed, reorg behavior around H_snap, and any compatibility concerns with existing node implementations.

[jonatack]

Thank you for your proposal, but this PR was opened prematurely:

• No dedicated prior post published to the bitcoin-dev mail list or to Delving Bitcoin to present your idea (and eventually later, a draft proposal) and have prior discussion of concept, technical merit, or soundness before opening this PR.
• If neither forum accepts your publication, please see it as feedback to improve rather than censorship.
• This is yet another confiscation proposal (along with an AI-written one yesterday at #2017 (comment)).
• This one appears to introduce a selective, rigid, brittle confiscation tool, attack bitcoin's fungibility via content-based rules -- a slippery slope, even if justified or narrow--, and lower the bar for future misuse or censorship attacks on bitcoin, e.g. by reusing this to blacklist any UTXO set, like mixers or sanctioned coins.
• Thus, I would expect this to be highly controversial, to say the least, and it certainly needs community discussion before being proposed here.
• The author's GitHub was created yesterday, and its sole action recently has been to open this PR.
• This appears to be mostly-to-completely LLM-generated output, which, given the absence of history of proof of work or contribution shown by the author (previous point), is not a good sign and does not encourage experienced contributors to spend time reviewing this.

As such, this appears to be a spamming or gaming of the BIPs process. Please don't waste the community's time this way. My suggestion would be to show proof of work and contribution over time and follow the BIP 2 process. Thank you.