set image version to 0.4.2 in helm chart (#160)

Update model-management.md

Remove GKE node selector from default L4 profile (#147)

Fixes #146

Improve naming for helm chart releases (#143)

add helm linting and releasing (#142)

Add nomic embedding model and add model feature filtering (#130)

Add Nomic Embedding model:

> nomic-embed-text is a large context length text encoder that surpasses
OpenAI text-embedding-ada-002 and text-embedding-3-small performance on
short and long context tasks.

 See: https://ollama.com/library/nomic-embed-text

Also: Add query string filtering in the model listing endpoint to avoid
OpenWebUI adding embedding models in the dropdown list of chat-able
models.

```go
// Example (default):  /v1/models
// Example (single):   /v1/models?feature=TextEmbedding
// Example (multiple): /v1/models?feature=TextGeneration&feature=TextEmbedding
```

run unit tests on PRs (#127)

Recreate subscription on receive message error and add pubsub e2e test (

#116)

Co-authored-by: Nick Stogner <[email protected]>

fix autoscaler stats aggregation (#114)

The stats were incorrectly being summed together. This was causing Lingo
to not scale up as high as it should when there are multiple lingo
replicas.

Add test cases for movingaverage (#110)