Extract rows and apply object embeddings (#42)

* - Restructured the extract directories
- Added an extractor for 'rows' == a row of a table
- Added a row extractor prompt to prompter.
* Add row support to template prompter
* Row extraction working
* Bump version
* Emit extracted info
* Object embeddings store
* Invocation script
* Add script to package, remove cruft output
* Write rows to Cassandra
* Remove output cruft

tests/test-rows-prompt

@@ -0,0 +1,51 @@
+#!/usr/bin/env python3
+
+import pulsar
+from trustgraph.clients.prompt_client import PromptClient
+from trustgraph.objects.object import Schema
+from trustgraph.objects.field import Field, FieldType
+
+schema = Schema(
+    name="actors",
+    description="actors in this story",
+    fields=[
+        Field(
+            name="name", type=FieldType.STRING,
+            description="Name of the animal or person in the story"
+        ),
+        Field(
+            name="legs", type=FieldType.INT,
+            description="Number of legs of the animal or person"
+        ),
+        Field(
+            name="notes", type=FieldType.STRING,
+            description="Additional notes or observations about this animal or person"
+        ),
+    ]
+)
+
+chunk = """I noticed a cat in my garden.  It is a four-legged animal
+which is a mammal and can be tame or wild.  I wonder if it will be friends
+with me?  I think the cat's name is Fred and it has 4 legs.
+There is also a dog barking outside.  The dog has 4 legs also.
+The dog comes to my call when I shout "Come here, Bernard".
+
+I am also standing in the garden, my name is Steve and I have 2 legs.
+
+My friend Clifford is coming to visit shortly, he has 3 legs due to
+a freak accident at birth.
+"""
+
+p = PromptClient(pulsar_host="pulsar://localhost:6650")
+
+resp = p.request_rows(
+    schema=schema,
+    chunk=chunk,
+)
+
+for d in resp:
+    print(f"Name: {d['name']}")
+    print(f"  No. of legs: {d['legs']}")
+    print(f"  Notes: {d['notes']}")
+    print()
+