chore: initialize EverOS 1.0.0

md-first memory extraction framework for AI agents.

Markdown is the single source of truth; SQLite holds state and LanceDB
provides the rebuildable vector + BM25 + scalar index. The codebase follows
a single-direction DDD layering (entrypoints -> service -> memory -> infra,
with component / core / config cross-cutting) enforced by import-linter.

Engineering surface:
- Coding conventions in .claude/rules/ (path-scoped) and workflows in
  .claude/skills/ (/commit, /new-branch, /pr).
- GitHub Actions CI runs make lint + test + integration; pre-commit mirrors
  the gates locally (ruff, hygiene hooks, gitlint commit-msg).
- Commit messages follow Conventional Commits, enforced by gitlint.
- make lint also enforces datetime two-zone discipline and OpenAPI drift.

tests/unit/test_memory/test_search/conftest.py

@@ -0,0 +1,27 @@
+"""Shared fixtures for ``memory.search`` unit tests.
+
+The project default is ``EVEROS_SEARCH__VECTOR_STRATEGY=maxsim_atomic`` —
+that path queries both the ``atomic_fact`` table and the ``episode`` table
+to do MaxSim. The existing VECTOR-route tests in ``test_manager.py`` were
+written against the legacy single-vector ``episode`` path and stub only the
+episode recaller (atomic_fact recaller is a no-data stub).
+
+Force the legacy ``episode`` strategy by default for these tests so they
+keep asserting against the dense-recall path they were designed to cover.
+MaxSim-specific tests opt back into ``maxsim_atomic`` by overriding the env
+var inside their own body.
+"""
+
+from __future__ import annotations
+
+import pytest
+
+from everos.config.settings import load_settings
+
+
+@pytest.fixture(autouse=True)
+def _force_episode_vector_strategy(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setenv("EVEROS_SEARCH__VECTOR_STRATEGY", "episode")
+    load_settings.cache_clear()
+    yield
+    load_settings.cache_clear()