docs(contracts): record dust-position entry_time gaming constraint

[pheobeayo]

Summary

• Documents a known gaming vector in Entry(address)/get_entry_time(): deposit() only stamps a fresh entry time when Balance(caller) was 0, and withdraw() only clears it on a full exit — so leaving a dust balance after a partial withdrawal lets a much later, much larger top-up inherit the original day-one timestamp instead of getting a fresh one.
• No code change: entry_time currently has zero functional consequence beyond display, so there's nothing to fix yet. This records the constraint now so it can't be silently missed if a duration-gated feature (fee discount, loyalty multiplier, vesting) is ever built against the raw value later.
• Adds guidance under get_entry_time() that any such feature should read a size-weighted average timestamp on top-up rather than raw entry_time, so a large late deposit meaningfully pulls the entry time forward.
• Docs-only change to apps/docs/architecture/vault-contract.md, the canonical reference docs/contracts.md already points readers to for this contract.

Test plan

• pnpm lint && pnpm typecheck && pnpm test pass locally
• Docs-only change — no contract or test behavior affected; verified the note is placed under both deposit() and get_entry_time() so it's visible from either entry point

Closes #577

[vercel]

@pheobeayo is attempting to deploy a commit to the Collins' projects Team on Vercel.

A member of the Team first needs to authorize it.

[drips-wave]

@pheobeayo 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

[collinsezedike]

The content here is accurate, both notes check out against the actual contract behavior. One inline fix needed before this can merge.

[collinsezedike]

This file is missing its trailing newline at the end. This isn't just style, it'll fail Lint & Typecheck since pnpm format:check (Prettier) enforces a trailing newline and this file isn't excluded in .prettierignore. Add the newline back and this is good to go.

[collinsezedike]

@pheobeayo this PR also has merge conflicts against main now, please rebase and resolve those alongside the newline fix above.

[pheobeayo]

@pheobeayo this PR also has merge conflicts against main now, please rebase and resolve those alongside the newline fix above.

fixed!

[collinsezedike]

@pheobeayo I fixed the trailing-newline issue directly and restored a sentence the dust-gaming edit had accidentally dropped (transfer-out addresses still self-heal to 0 on next read, that's a separate case from the dust constraint this PR adds). Merging now.