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/parser/test_document_reader.py

@@ -11,8 +11,8 @@ from typing import Any, Dict, List
 
 import pytest
 
-import application.parser.document_reader as dr
-from application.parser.document_reader import (
+import docsgpt.parser.document_reader as dr
+from docsgpt.parser.document_reader import (
     bound_parse_payload,
     parse_document_bytes,
     truncate_text_head_tail,
@@ -229,7 +229,7 @@ def test_default_markdown_with_tables_converts_docling_once(monkeypatch):
     # engine=auto resolves to the server default; this test is about the docling path.
     monkeypatch.setattr(dr.settings, "DOC_PARSER_ENGINE", "docling")
 
-    from application.parser.file.docling_parser import DoclingParser
+    from docsgpt.parser.file.docling_parser import DoclingParser
 
     counter = {"instances": 0, "converts": 0}
 
@@ -546,7 +546,7 @@ def test_vanilla_converter_applies_inference_settings(monkeypatch, tmp_path):
 
     called = []
     monkeypatch.setattr(
-        "application.parser.file.docling_parser._apply_inference_settings",
+        "docsgpt.parser.file.docling_parser._apply_inference_settings",
         lambda: called.append(True),
     )
 
@@ -618,7 +618,7 @@ def test_structured_tables_render_blank_cells_as_strings(monkeypatch, tmp_path):
             return types.SimpleNamespace(document=_Doc())
 
     monkeypatch.setattr(
-        "application.parser.file.docling_parser._apply_inference_settings",
+        "docsgpt.parser.file.docling_parser._apply_inference_settings",
         lambda: None,
     )
     fake_docling = types.ModuleType("docling")
@@ -753,7 +753,7 @@ def test_tables_are_not_collected_when_the_pdf_parser_is_not_docling(monkeypatch
     """Under OCR_BACKEND=native the docling engine hands PDFs to the native OCR
     parser; a vanilla DocumentConverter table pass would OCR the scan again."""
 
-    from application.parser.file.ocr_parser import NativeOcrPdfParser
+    from docsgpt.parser.file.ocr_parser import NativeOcrPdfParser
 
     native = NativeOcrPdfParser()
     native._parser_config = {}
@@ -777,8 +777,8 @@ def test_tables_survive_native_ocr_delegation_for_text_only_pdfs(monkeypatch):
     to that parser, so tables must ride that single conversion instead of being
     dropped by the scan exclusion."""
 
-    from application.parser.file.docling_parser import DoclingParser
-    from application.parser.file.ocr_parser import NativeOcrPdfParser
+    from docsgpt.parser.file.docling_parser import DoclingParser
+    from docsgpt.parser.file.ocr_parser import NativeOcrPdfParser
 
     class _Docling(DoclingParser):
         def __init__(self):