feat(processing_engine): Runtime and Write-back improvements

[jacksonrnewhouse]

This makes two major changes, running plugins in their own tasks and enabling writing back line protocol to the WriteBuffer.

Running Plugins as Separate Tasks

Each plugin now runs in its own task, started either when created through the API or when the system is started. It is connected to the rest of the system through two mechanisms:

• A broadcast channel whose sender is owned by the QueryableBuffer and whose receiver it polls from.
• An Arc to enable write-backs. This also gives access to the catalog and much of the rest of the db as we continue to build out the API.
  Right now the only message being sent is Arc<WalContents>, but I expect the PluginEvent enum to be expanded both for other triggering events and control messages, e.g. Shutdown, Status, etc.

Write-Back Line Protocol

The running pythons have a reference to the WriteBuffer, and this PR also adds a simple receiver for python to write line protocols to with the insert_line_protocol call. Everything is buffered per PluginEvent, and data is copied several times, so there is a lot of room for improvement.

Example REST calls

Creating a Plugin:

curl -X POST 'http://127.0.0.1:8181/api/v3/configure/processing_engine_plugin' \ 
-H 'Content-Type: application/json' \
-d '{
    "db": "load_test",
    "plugin_name": "counter_output",
    "plugin_type": "wal_rows",                 
    "function_name": "count_points",
    "code": "import time\nfrom collections import defaultdict\nfrom influxdb_client_3.write_client.client.write.point import Point\n\ndef count_points(iterator, output):\n    start_time = time.time()\n    table_counts = defaultdict(int)\n    \n    while True:\n        point = iterator.next_point()\n        if point is None:\n            break\n            \n        measurement = point._name\n        table_counts[measurement] += 1\n    \n    for table_name, count in table_counts.items():\n        stats_point = Point(\"table_counts\")\n        stats_point.tag(\"table_name\", table_name)\n        stats_point.field(\"row_count\", count)\n        output.insert_line_protocol(str(stats_point))\n    \n    end_time = time.time()\n    duration = end_time - start_time\n    \n    print(f\"Processing completed in {duration:.2f} seconds\")"
}'

Creating a Trigger:

 curl -X POST 'http://127.0.0.1:8181/api/v3/configure/processing_engine_trigger' \
-H 'Content-Type: application/json' \
-d '{
    "db": "load_test",
    "plugin_name": "counter_output",
    "trigger_name": "counter_output_trigger",
    "trigger_specification": {
         "single_table_wal_write": {"table_name": "measurement_data"}
    }
}'

[pauldix]

A couple of small comments, but otherwise LGTM