refactor: rename the application package to docsgpt

The backend import package is now docsgpt, the name it will carry on PyPI;
application was far too generic to install into anyone's site-packages.
git mv plus a mechanical rewrite of every import, dotted string and path
reference: 734 Python files, the compose files, Dockerfile, workflows, docs,
setup scripts, devcontainer, k8s manifests, vscode config, pytest and coverage
config, .gitignore. Behaviour is unchanged.

Kept for one release:
- A top-level application package whose meta-path finder resolves
  application.x.y to the already-imported docsgpt.x.y object, so old imports
  and entry points (celery -A application.app.celery,
  uvicorn application.asgi:asgi_app) keep working with a FutureWarning.
- Celery registers every application.* task name as an alias of its
  docsgpt.* task on start-up, so messages queued by the previous release still
  run. The redbeat key prefix moves to redbeat:docsgpt:v2: so schedule entries
  the previous release wrote are left unread instead of firing twice.

The backend image builds from the repository root (docker build -f
docsgpt/Dockerfile .) so it can ship the alias package; a root .dockerignore
allow-lists docsgpt/ and application/ and keeps caches, local data, .env
files, the sample index files and the Dockerfile out. Compose and the image
workflows point at the new context.

tests/graphrag/test_extraction.py

@@ -13,10 +13,10 @@ import uuid
 
 import pytest
 
-import application.graphrag.extraction as extraction_module
-from application.graphrag.store import GraphStore
-from application.storage.db.source_config import SourceConfig
-from application.vectorstore import pgconn
+import docsgpt.graphrag.extraction as extraction_module
+from docsgpt.graphrag.store import GraphStore
+from docsgpt.storage.db.source_config import SourceConfig
+from docsgpt.vectorstore import pgconn
 
 extract_graph_for_source = extraction_module.extract_graph_for_source
 
@@ -55,7 +55,7 @@ def _live_store(monkeypatch, info):
     dsn = _ephemeral_dsn(info)
     # The pipeline builds its own GraphStore() from settings, so point those at
     # the ephemeral cluster too — never at the operator's configured DB.
-    from application.core import settings as settings_module
+    from docsgpt.core import settings as settings_module
 
     monkeypatch.setattr(
         settings_module.settings, "PGVECTOR_CONNECTION_STRING", dsn, raising=False
@@ -103,7 +103,7 @@ class _StubEmbedding:
 
 @pytest.fixture
 def stub_embedding(monkeypatch):
-    from application.core.settings import settings
+    from docsgpt.core.settings import settings
 
     # The resolver short-circuits to the remote API when this is configured,
     # which would bypass the stub on a dev machine that sets it.
@@ -436,7 +436,7 @@ class TestEmbeddingsResolution:
         fake_store = MagicMock()
         fake_store.pending_chunks.return_value = []
         monkeypatch.setattr(
-            "application.graphrag.store.GraphStore", lambda *a, **k: fake_store
+            "docsgpt.graphrag.store.GraphStore", lambda *a, **k: fake_store
         )
         _install_stub_llm(monkeypatch, _StubLLM([]))
 

tests/graphrag/test_graphrag_available.py

@@ -4,8 +4,8 @@ from __future__ import annotations
 
 import pytest
 
-from application.graphrag import graphrag_available
-from application.core.settings import settings
+from docsgpt.graphrag import graphrag_available
+from docsgpt.core.settings import settings
 
 
 @pytest.mark.unit

tests/graphrag/test_retriever_live.py

@@ -15,8 +15,8 @@ from unittest.mock import MagicMock, Mock, patch
 
 import pytest
 
-from application.retriever.graph_rag import GraphRAGRetriever
-from application.vectorstore import pgconn
+from docsgpt.retriever.graph_rag import GraphRAGRetriever
+from docsgpt.vectorstore import pgconn
 
 pytestmark = pytest.mark.integration
 
@@ -64,12 +64,12 @@ def _close_pools():
 def stub_embeddings():
     stub = _StubEmbeddings()
     with patch(
-        "application.vectorstore.base.get_embeddings", return_value=stub
+        "docsgpt.vectorstore.base.get_embeddings", return_value=stub
     ), patch(
-        "application.vectorstore.base.BaseVectorStore._get_embeddings",
+        "docsgpt.vectorstore.base.BaseVectorStore._get_embeddings",
         return_value=stub,
     ), patch(
-        "application.retriever.graph_rag.get_embeddings", return_value=stub
+        "docsgpt.retriever.graph_rag.get_embeddings", return_value=stub
     ):
         yield stub
 
@@ -86,7 +86,7 @@ def live_dsn(postgresql, monkeypatch):
         pytest.skip(f"pgvector extension unavailable: {exc}")
 
     dsn = _dsn(postgresql.info)
-    from application.core import settings as settings_module
+    from docsgpt.core import settings as settings_module
 
     settings = settings_module.settings
     monkeypatch.setattr(settings, "VECTOR_STORE", "pgvector", raising=False)
@@ -99,8 +99,8 @@ def live_dsn(postgresql, monkeypatch):
 
 def _seed(dsn, graph_source_id):
     """Ingest one chunk and one graph node for ``graph_source_id``."""
-    from application.graphrag.store import GraphStore
-    from application.vectorstore.pgvector import PGVectorStore
+    from docsgpt.graphrag.store import GraphStore
+    from docsgpt.vectorstore.pgvector import PGVectorStore
 
     vector_store = PGVectorStore(
         source_id=graph_source_id, connection_string=dsn
@@ -132,7 +132,7 @@ def _seed(dsn, graph_source_id):
 
 def _retriever(sources):
     with patch(
-        "application.retriever.classic_rag.LLMCreator.create_llm",
+        "docsgpt.retriever.classic_rag.LLMCreator.create_llm",
         Mock(return_value=MagicMock()),
     ):
         return GraphRAGRetriever(
@@ -194,7 +194,7 @@ class TestGraphRAGRetrieverLive:
     def test_all_graphless_sources_take_one_classic_call(
         self, live_dsn, stub_embeddings
     ):
-        from application.graphrag.store import GraphStore
+        from docsgpt.graphrag.store import GraphStore
 
         store = GraphStore(connection_string=live_dsn)
         store._ensure_tables()  # tables exist, but no source has a graph

tests/graphrag/test_store.py

@@ -20,9 +20,9 @@ from unittest.mock import MagicMock, patch
 
 import pytest
 
-import application.graphrag.store as store_module
-from application.vectorstore import pgconn
-from application.vectorstore import pgvector as pgvector_module
+import docsgpt.graphrag.store as store_module
+from docsgpt.vectorstore import pgconn
+from docsgpt.vectorstore import pgvector as pgvector_module
 
 GraphStore = store_module.GraphStore
 
@@ -374,13 +374,13 @@ class TestGraphStoreLive:
 
     def test_the_vector_store_reuses_the_graph_store_pool(self, store, postgresql):
         """Same DSN, one pool: the graph store does not double the connections."""
-        from application.vectorstore.pgvector import PGVectorStore
+        from docsgpt.vectorstore.pgvector import PGVectorStore
 
         stub = MagicMock()
         stub.dimension = TEST_EMBEDDING_DIM
         stub.embed_query.return_value = [0.0] * TEST_EMBEDDING_DIM
         with patch(
-            "application.vectorstore.base.BaseVectorStore._get_embeddings",
+            "docsgpt.vectorstore.base.BaseVectorStore._get_embeddings",
             return_value=stub,
         ):
             vector_store = PGVectorStore(
@@ -462,7 +462,7 @@ class TestGraphStoreParameterization:
         assert params == (embedding, sid, embedding, 5)
 
     def test_graph_overview_binds_source_and_clamps_limit(self):
-        from application.graphrag.store import GRAPH_OVERVIEW_MAX_LIMIT
+        from docsgpt.graphrag.store import GRAPH_OVERVIEW_MAX_LIMIT
 
         store, cursor = self._store_with_mock_conn()
         cursor.fetchall.return_value = []
@@ -499,7 +499,7 @@ class TestEmbeddingDim:
     """The graph table dimension is derived from the configured model (FIX 1)."""
 
     def test_uses_configured_model_dimension(self, monkeypatch):
-        from application.vectorstore import base as base_module
+        from docsgpt.vectorstore import base as base_module
 
         monkeypatch.setattr(base_module.settings, "EMBEDDINGS_BASE_URL", None)
         fake_embedding = MagicMock()
@@ -520,7 +520,7 @@ class TestEmbeddingDim:
         ``getattr`` with a default cannot catch that -- the attribute exists --
         so the width reached the DDL as ``vector(None)``.
         """
-        from application.vectorstore import base as base_module
+        from docsgpt.vectorstore import base as base_module
 
         monkeypatch.setattr(base_module.settings, "EMBEDDINGS_BASE_URL", None)
         fake_embedding = MagicMock()
@@ -536,7 +536,7 @@ class TestEmbeddingDim:
         assert store._embedding_dim() == store_module.DEFAULT_NAME_EMBEDDING_DIM
 
     def test_falls_back_to_default_dimension(self, monkeypatch):
-        from application.vectorstore import base as base_module
+        from docsgpt.vectorstore import base as base_module
 
         monkeypatch.setattr(base_module.settings, "EMBEDDINGS_BASE_URL", None)
         fake_embedding = object()
@@ -562,7 +562,7 @@ class TestEmbeddingDimResolution:
         monkeypatch.setattr(GraphStore, "_embedding_dim", _REAL_EMBEDDING_DIM)
 
         with patch(
-            "application.vectorstore.base.get_embeddings",
+            "docsgpt.vectorstore.base.get_embeddings",
             return_value=fake_embedding,
         ) as mock_resolver:
             store = GraphStore.__new__(GraphStore)