Usage aggregation via Dictionary<string, long>

[SteveSandersonMS]

This is a much simpler version of #5707 in case we prefer this approach. It limits additional usage data to be of type long, with the payoff that we require much less implementation and new public API surface.

If we think this will suffice, it's certainly advantageous vs #5707. Arguably we could proceed with this as-is, and then in the future if further "additional usage" value types are needed, we'd still have the option to add some other property called AdditionalValues (or similar) to UsageDetails which accepts a wider range of types.

Microsoft Reviewers: Open in CodeFlow

[dotnet-comment-bot]

🎉 Good job! The coverage increased 🎉
Update MinCodeCoverage in the project files.

Project	Expected	Actual
Microsoft.Extensions.AI	83	84
Microsoft.Extensions.Diagnostics.Probes	70	76
Microsoft.Extensions.Caching.Hybrid	75	86

Full code coverage report: https://dev.azure.com/dnceng-public/public/_build/results?buildId=883335&view=codecoverage-tab

[dotnet-comment-bot]

🎉 Good job! The coverage increased 🎉
Update MinCodeCoverage in the project files.

Project	Expected	Actual
Microsoft.Extensions.Diagnostics.Probes	70	76
Microsoft.Extensions.Caching.Hybrid	75	86
Microsoft.Extensions.AI.AzureAIInference	83	91
Microsoft.Extensions.AI	83	84
Microsoft.Extensions.AI.Abstractions	83	84

Full code coverage report: https://dev.azure.com/dnceng-public/public/_build/results?buildId=885946&view=codecoverage-tab

[dotnet-comment-bot]

🎉 Good job! The coverage increased 🎉
Update MinCodeCoverage in the project files.

Project	Expected	Actual
Microsoft.Extensions.Diagnostics.Probes	70	76
Microsoft.Extensions.Caching.Hybrid	75	86
Microsoft.Extensions.AI.AzureAIInference	83	91
Microsoft.Extensions.AI.Abstractions	83	84
Microsoft.Extensions.AI	83	84

Full code coverage report: https://dev.azure.com/dnceng-public/public/_build/results?buildId=886016&view=codecoverage-tab