feat: add full-rate and trickle SP testing tiers

[silent-cipher]

Part of #681

Summary

• Classify eligible storage provider into full-rate and trickle testing tiers.
• Keep approved providers and configured expected-approval candidates at the network-configured rates.
• Limit other eligible provider to 1 deal and 1 data-set creation attempt every 4 hours, with a target of 1 dataset.
• Add per-network expected-approved providerID and address configuration.
• Apply the tier specific dataset target during scheduling, provisioning, and deal dataset selection.
• Schedule dataset lifecycle checks only for full-rate providers, remove stale lifecycle schedules after a downgrade, and recheck eligibility when queued lifecycle jobs execute.
• Leave retrieval, sample retrieval, pull-check, and cleanup cadences unchanged.

USE_ONLY_APPROVED_PROVIDERS remain the outer provider-eligibility gate. when enabled, expected-approved provider lists do not admit unapproved providers.

Out of scope

• Terminating existing datasets or payment rails above the new trickle target. This requires a dedicated on-chain termination job.
• Changing pull-check scheduling. Pull-check signing is used for authorization validation and does not itself submit the wallet-spending contract call.
• Cancelling or skipping the general deal jobs that were queued before a tier downgrade.

[Copilot]

Pull request overview

Implements provider testing tiering to reduce wallet-spend exposure from newly registered/ephemeral storage providers by running only “full-rate” providers at configured rates and throttling all other eligible providers to a fixed “trickle” rate per network.

Changes:

• Adds per-network “expected approved” allowlists (IDs + addresses) and a shared isFullRateTier helper to classify providers into full-rate vs trickle tiers.
• Applies tier-specific scheduling and dataset targeting (deals + dataset creation rates, min dataset targets), and restricts dataset lifecycle checks to full-rate providers (including stale-schedule cleanup and runtime re-eligibility checks).
• Updates configuration loading/validation, docs, and examples to describe and support the new tiering knobs.

Reviewed changes

Copilot reviewed 16 out of 16 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
docs/environment-variables.md	Documents tiering behavior and adds expected-approved env vars.
apps/backend/.env.example	Adds example expected-approved vars and tiering notes.
apps/backend/src/config/constants.ts	Introduces fixed trickleTierRates constants.
apps/backend/src/config/env.schema.ts	Adds env validation for expected-approved lists.
apps/backend/src/config/loader.ts	Loads expected-approved ID/address lists into per-network config.
apps/backend/src/config/network-fields.ts	Registers expected-approved vars as per-network + chain-specific.
apps/backend/src/config/types.ts	Extends network config types with expected-approved sets.
apps/backend/src/common/sp-tier.ts	Adds helper to classify providers into full-rate tier.
apps/backend/src/providers/repositories/storage-provider.repository.ts	Extends active-provider projection to include isApproved for tiering.
apps/backend/src/providers/repositories/storage-provider.repository.spec.ts	Updates repository tests for new projection shape.
apps/backend/src/jobs/repositories/job-schedule.repository.ts	Adds deletion helper for per-address schedules (used to clear stale lifecycle schedules after downgrades).
apps/backend/src/jobs/jobs.service.ts	Applies tiering to scheduling + dataset targets; gates lifecycle schedules and re-checks tier at job execution.
apps/backend/src/jobs/jobs.service.spec.ts	Updates and adds tests for lifecycle scheduling behavior under tiering.
apps/backend/src/deal/deal.service.ts	Applies tier-specific min-dataset target to deal dataset selection.
apps/backend/src/deal/deal.service.spec.ts	Updates tests for the new minDataSets parameter path.
apps/backend/src/wallet-sdk/wallet-sdk.service.spec.ts	Updates base network config test fixtures for expected-approved fields.

Suppressed comments (2)

apps/backend/src/jobs/jobs.service.spec.ts:1543

• The findActiveAddresses repository contract includes providerId, and ensureScheduleRows passes it into isSpBlocked(...) / tiering. This test mock omits providerId, which can hide regressions in ID-based blocklisting and diverges from the runtime shape.

    providerRegistryRepositoryMock.findActiveAddresses.mockResolvedValueOnce([{ address: "0xaaa", isApproved: true }]);

apps/backend/src/jobs/jobs.service.spec.ts:1569

• The findActiveAddresses repository contract includes providerId, and ensureScheduleRows passes it into isSpBlocked(...) / tiering. This test mock omits providerId, which can hide regressions in ID-based blocklisting and diverges from the runtime shape.

    providerRegistryRepositoryMock.findActiveAddresses.mockResolvedValueOnce([{ address: "0xaaa", isApproved: false }]);

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[SgtPooki]

in general lgtm, but I left a few comments

[silent-cipher]

I'll merge this after https://github.com/FilOzone/infra/pull/367