docs: design Soroban event topic strategy for Stellar announcer

[David-patrick-chuks]

Overview

This PR adds the Stellar design note for issue #23 and recommends a new Soroban event topic strategy for stealth-announcer so indexers can filter announcements server-side via getEvents. The proposal moves toward a four-topic indexed schema centered on scheme_id and view_tag_bucket, quantifies the expected scan-bandwidth savings, and documents the privacy trade-off explicitly.

Related Issue

Closes #23

Changes

⚙️ Event Topic Design

• [ADD] stellar/EVENT_TOPIC_DESIGN.md
• Added a design doc for a new indexed Soroban event topic schema.
• Proposed topic layout: ("announce", scheme_id, view_tag_bucket, metadata_kind).
• Recommended moving stealth_address, ephemeral_pub_key, and metadata into event data.
• Documented why stealth_address should no longer consume an indexed topic slot.
• Included expected client query shapes for Stellar RPC getEvents filtering.

📊 Savings Model

• Added a back-of-the-envelope model for a mainnet scenario with 50 announcements/sec and 10,000 scanners.
• Quantified the expected improvement from bucketed filtering at roughly 256x less scan traffic.
• Estimated aggregate savings at about 21.5 TB/day of response traffic avoided.

🔐 Privacy + Compatibility Analysis

• Documented the privacy cost of exposing a one-byte view_tag_bucket in public topics.
• Compared the two rollout paths discussed in the issue.
• Recommended path A: deploy a new v2 announcer with scheme_id = 2 and let SDKs read both versions during migration.

📝 Docs

• [MODIFY] README.md
• Added a pointer to stellar/EVENT_TOPIC_DESIGN.md in the Stellar section.

Verification Results

Acceptance Criteria	Status
Design doc at stellar/EVENT_TOPIC_DESIGN.md	✅
Quantified savings model included	✅
Recommendation is explicit	✅
Privacy trade-off is explicit and unambiguous	✅
Follow-up issue filed for contract change	✅
Follow-up issue filed for SDK fetch change	✅
Stellar workspace tests pass	✅

Screenshots / Command Results

cargo test --workspace ✅ passed

[drips-wave]

@David-patrick-chuks Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits

[truthixify]

This is the kind of design doc I want to see — a clean problem statement, the 256× back-of-envelope, and the migration path argued in both directions before landing on a recommendation. The "Why stealth_address should leave topics" section is sharp; the 4-slot budget framing makes the trade-offs obvious.

Good call filing #24 and #25 as the contract + SDK follow-ups. Merging.

Thanks @David-patrick-chuks.