Add commitment discount eligibility open data set

[RolandKrummenacher]

Summary

Adds a new open data CSV (CommitmentDiscountEligibility.csv) listing Azure meters eligible for Reserved Instances and/or Savings Plans, sourced from the Azure Retail Prices API.

• src/scripts/Update-CommitmentDiscountEligibility.ps1 — PowerShell script that pages through the Retail Prices API to collect RI-eligible meters (priceType eq 'Reservation') and SP-eligible meters (Consumption items with savingsPlan arrays), then merges them into a single CSV.
• .github/workflows/opendata-commitment-eligibility.yml — GitHub Action that runs monthly (1st of each month) or on manual dispatch, executes the script, and creates a PR if data changed.
• src/open-data/CommitmentDiscountEligibility.csv — Initial dataset with ~84K meters.

Motivation

FinOps Hub Prices_v1_2() / Prices_v1_0() functions currently compute commitment discount eligibility via cross-row lookups at query time, scanning the full prices table on every call. This open data set provides a pre-computed lookup that can replace those per-query scans, and also solves the per-batch ingestion issue described in #1625.

Schema

Column	Description
MeterId	Azure meter GUID (region-specific)
ServiceName	Azure service name (informational)
ServiceFamily	Service family (informational)
ReservationEligible	Whether the meter has RI pricing
SavingsPlanEligible	Whether the meter has SP pricing

Scope

• Covers Reserved Instances and Savings Plans only
• Pre-Purchase Plans (e.g., Databricks, Sentinel SCUs) show as Reservation in the API and are included
• Commitment tiers (e.g., Sentinel GB/day, Log Analytics) are out of scope for now — they use a different pricing model

Next steps

• Integrate into Build-OpenData.ps1 to generate PowerShell functions and KQL lookup tables
• Use in Hub ingestion/view functions to replace cross-row eligibility computation

Test plan

• Full test run completed — script fetched ~142 pages of RI data and ~609 pages of Consumption data
• Rate limit retry logic validated (exponential backoff with up to 5 retries)
• Output CSV verified: 84,612 unique meters, no duplicates
• Review GitHub Action workflow triggers and PR creation logic
• Validate a sample of meter IDs against the Azure Pricing Calculator

🤖 Generated with Claude Code