refactor(memory): remove legacy memory v1

[qin-ctx]

Description

Remove the legacy memory v1 implementation now that memory v2 is the only supported runtime path.

Related Issue

N/A

Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Refactoring (no functional changes)
• Performance improvement
• Test update

Changes Made

• Delete legacy v1 compressor, extractor, deduplicator modules, v1 compression prompt templates, and v1-only tests.
• Make memory configuration and create_session_compressor v2-only, rejecting memory.version = "v1" explicitly.
• Update docs, prompt guide references, sample dataset text, and remaining tests to point at the v2 memory path.
• Remove tests that referenced the missing SchemaPromptGenerator helper instead of adding that helper to production code in this cleanup PR.

Testing

• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• I have tested this on the following platforms:
  • Linux
  • macOS
  • Windows

Commands run:

• .venv/bin/python -m pytest -q --no-cov tests/test_config_loader.py tests/misc/test_semantic_config.py (38 passed)
• .venv/bin/python -m compileall -q openviking/session openviking/service openviking_cli/utils/config tests/test_config_loader.py tests/misc/test_semantic_config.py tests/session/test_commit_skill_inference.py tests/session/memory/test_memory_react.py
• git diff --check
• rg scan for legacy memory v1 modules/classes/prompts and SchemaPromptGenerator references (no matches)

Screenshots (if applicable)

N/A

Additional Notes

A broader run including tests/session/test_commit_skill_inference.py is currently blocked by an existing main-branch LocalClient abstract method gap: LocalClient does not implement BaseClient.batch_add_messages. This PR intentionally does not add production code for that unrelated test setup issue.

[github-actions]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪

🏅 Score: 85

🧪 PR contains tests

🔒 No security concerns identified

✅ No TODO sections

🔀 Multiple PR themes Sub-PR theme: Add LocalClient.batch_add_messages implementation Relevant files: openviking/client/local.py Sub-PR theme: Add SchemaPromptGenerator for memory schema prompt generation Relevant files: openviking/session/memory/schema_model_generator.py

⚡ Recommended focus areas for review Potential regression: create_session_compressor no longer reads config for memory_version The create_session_compressor function no longer reads the memory.version from the config when memory_version is None. While the config validator now rejects v1, this changes the previous behavior where the config was the source of truth for the memory version. def create_session_compressor( vikingdb: VikingDBManager, memory_version: Optional[str] = None, skill_processor=None, ) -> "SessionCompressorV2": """ Create the v2 session compressor. Args: vikingdb: VikingDBManager instance memory_version: Deprecated optional override. Only "v2" is supported. Returns: SessionCompressorV2 instance """ if memory_version not in (None, "v2"): raise ValueError("memory.version only supports 'v2'; legacy memory v1 has been removed") logger.info("Using v2 memory compressor (templating system)") from openviking.session.compressor_v2 import SessionCompressorV2 return SessionCompressorV2(vikingdb=vikingdb, skill_processor=skill_processor) Type safety: registry_or_schemas is typed as Any The SchemaPromptGenerator.init accepts registry_or_schemas as Any, which reduces type safety. Consider using a Protocol or Union of expected types for better type checking. def __init__( self, registry_or_schemas: Any, template_context: Optional[Dict[str, Any]] = None, ):

[github-actions]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
General	Replace quadratic schema lookup with dictionary for O(n) performance Fix quadratic loop in SchemaPromptGenerator by pre-building a memory-type-to-schema lookup dictionary. The current implementation iterates over all schemas for every schema (O(n²)) when generating type descriptions, which can be optimized to O(n) using a dict. openviking/session/memory/schema_model_generator.py [359-401] def __init__( self, registry_or_schemas: Any, template_context: Optional[Dict[str, Any]] = None, ): if hasattr(registry_or_schemas, "list_all"): self.schemas = list(registry_or_schemas.list_all(include_disabled=True)) else: self.schemas = list(registry_or_schemas or []) self._template_context = dict(template_context or {}) + self._schema_by_type = {schema.memory_type: schema for schema in self.schemas} def _render(self, text: str) -> str: if not text: return text if "{{" not in text and "{%" not in text and "{#" not in text: return text return TemplateUtils.render(text, self._template_context, strip=False) def generate_type_descriptions(self) -> str: lines = ["## Available Memory Types"] for schema in self.schemas: lines.append("") lines.append(f"### {schema.memory_type}") if schema.description: lines.append(self._render(schema.description)) field_descriptions = self.generate_field_descriptions(schema.memory_type) if field_descriptions: lines.append(field_descriptions) return "\n".join(lines).strip() def generate_field_descriptions(self, memory_type: str) -> Optional[str]: - schema = next((item for item in self.schemas if item.memory_type == memory_type), None) + schema = self._schema_by_type.get(memory_type) if schema is None: return None lines = [f"### {schema.memory_type} Fields"] for field in schema.fields: field_type = getattr(field.field_type, "value", field.field_type) merge_op = getattr(field.merge_op, "value", field.merge_op) description = self._render(field.description) suffix = f": {description}" if description else "" lines.append(f"- `{field.name}` ({field_type}, {merge_op}){suffix}") return "\n".join(lines) Suggestion importance[1-10]: 5 __ Why: The suggestion correctly identifies and fixes a quadratic loop in SchemaPromptGenerator by adding a memory-type-to-schema lookup dictionary, improving performance from O(n²) to O(n) for type descriptions generation. This is a valid optimization with moderate impact on code efficiency.	Low