Multiple customers are reporting duplicate records or missing answers,
most times because of an incorrect KM setup such as using a volatile
document storage with a persistent vector store (or viceversa).

This PR changes the memory builder behavior to detect these setups and
throw an exception in such cases. The exception can be suppressed if the
user needs such a setup, e.g. for testing purposes.

## Motivation and Context (Why the change? What's the scenario?)

When importing chat messages or running tests with small files or
without IDs, the storage might fill up with duplicate chunks of text
that affect the number of tokens used when generating an answer.

## High level description (Approach, Design)

When using the ASK API, after fetching N records from storage, discard
duplicates and use only unique chunks in the RAG prompt.
The behavior is configurable via
`SearchClientConfig.IncludeDuplicateFacts` and request context
`custom_rag_include_duplicate_facts_bool` arg.

Add basic configuration check to detect some warnings and errors. The
feature can be extended in future to check more services and more
details, e.g. availability and permissions.

How to use:
```
cd service/Service
dotnet run check
```

Updated the **OllamaSharp** package version in the
`Directory.Packages.props` file from 4.0.11 to 4.0.17.

See https://github.com/microsoft/kernel-memory/issues/963

Improve web client streaming performance

## Motivation and Context (Why the change? What's the scenario?)

Allow the use of [instance
metadata](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-instance-metadata.html)
for authentication and the proper use of [credential
chain](https://docs.aws.amazon.com/sdkref/latest/guide/standardized-credentials.html).

## High level description (Approach, Design)

Added AWS S3 Extension CredentialChain authentication method to use
default credentials

---------

Co-authored-by: Armando Portill <aportillo@fingercheck.com>
Co-authored-by: Devis Lucato <dluc@users.noreply.github.com>
Co-authored-by: Devis Lucato <devis@microsoft.com>

## Motivation and Context (Why the change? What's the scenario?)

Adds a new TokenUsage property to MemoryAnswer to hold information about token usage.

## High level description (Approach, Design)

* Include token count provided by the internal tokenizer
* Include token count provided by the service, if available
* Support streaming and multiple services, if needed

---------

Co-authored-by: Devis Lucato <dluc@users.noreply.github.com>