Scientific Metadata Search Engine client service using TypeSense

[Kezzsim]

We can successfully instantiate and connect to a new Typesense instance with this .yml file and the command tiled serve config config.yml

authentication:
  # The default is false. Set to true to enable any HTTP client that can
  # connect to _read_. An API key is still required to write.
  allow_anonymous_access: false
  single_user_api_key: "secret"  # for dev
trees:
  - path: /
    tree: catalog
    args:
      uri: "sqlite+aiosqlite:///:memory:"
      # or, uri: "sqlite+aiosqlite:////catalog.db"
      # or, "postgresql+asyncpg://..."
      writable_storage: "tmp/"
      init_if_not_exists: true
      typesense_client:
        api_key: "secret"
        schemas:
        nodes:
          - host: "http://localhost"
            port: 8108
            protocol: "http"

However there is still a lot of specific work to be done, namely items mentioned in #605 :

• Set up Typesense client dependencies
• Instantiate a client and have it connect with a typesense server
• Create a bluesky event model reader that generates a standard base schema
• Save schemas to files that are extensible and include-able by name so the entire schema does not have to be explicitly set each time.
  • (e.g. you could have a schema file for each beamline at NSLS-II with some of their metadata shape imparted)
• Hook into Context.__init__
  • Initialize Typesense instance by iterating over all data in the tiled instance
  • Upsert search engine data on database insertion by registering [after_insert] sqlalchemy hook