refactor(rag): simplify configuration and initialization

[ErikBjare]

• Reduce configuration complexity to just an enable flag
• Use default paths and collection names
• Add initialization state tracking
• Improve code organization and documentation

---

Important

Simplifies RAG configuration to a single enable flag, uses default paths, adds initialization tracking, and improves code organization and documentation.

• Configuration:
  • Simplified RAG configuration to a single enabled flag in gptme.toml.
  • Removed custom index_path and collection settings, using defaults instead.
• Initialization:
  • Added _init_run flag in rag.py to track initialization state and prevent redundant initializations.
• Code Organization:
  • Removed unused imports in _rag_context.py.
  • Improved logging for RAG availability and configuration status in rag.py.
• Documentation:
  • Updated RAG configuration and feature documentation in rag.py.

^{This description was created by for 8c2b370. It will automatically update as commits are pushed.}

[ellipsis-dev]

👍 Looks good to me! Reviewed everything up to dda0ec5 in 14 seconds

More details

• Looked at 151 lines of code in 3 files
• Skipped 0 files when reviewing.
• Skipped posting 3 drafted comments based on config settings.

1. gptme/tools/rag.py:17

• Draft comment:
  Removing configuration options like index_path and collection reduces flexibility. Consider allowing overrides for advanced users.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The PR aims to simplify the configuration and initialization of the RAG tool. However, the removal of configuration options like index_path, collection, and others might limit flexibility for users who need custom configurations. It's important to ensure that the default values are suitable for most use cases, but also consider providing a way to override them if needed.

2. gptme/tools/rag.py:109

• Draft comment:
  Ensure _init_run is thread-safe if used in a multi-threaded environment.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The PR introduces a global variable _init_run to track initialization state. This is a good practice to prevent re-initialization, but it should be ensured that this variable is thread-safe if the code is expected to run in a multi-threaded environment.

3. gptme/tools/_rag_context.py:74

• Draft comment:
  Consider allowing index_path and collection to be configurable for advanced users.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The RAGManager class in _rag_context.py has hardcoded default values for index_path and collection. While this simplifies the configuration, it may not be suitable for all users. Consider allowing these to be configurable through environment variables or a separate configuration file.

Workflow ID: wflow_ILqQDWor8sjiBJDE

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 72.37%. Comparing base (950c397) to head (8c2b370).
Report is 6 commits behind head on master.

✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #266      +/-   ##
==========================================
- Coverage   73.30%   72.37%   -0.94%     
==========================================
  Files          64       64              
  Lines        4211     4311     +100     
==========================================
+ Hits         3087     3120      +33     
- Misses       1124     1191      +67     

Flag	Coverage Δ
anthropic/claude-3-haiku-20240307	71.09% <100.00%> (-1.31%)	⬇️
openai/gpt-4o-mini	70.98% <100.00%> (-1.29%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

---

🚨 Try these New Features:

• JS Bundle Analysis - Avoid shipping oversized bundles

[ellipsis-dev]

👍 Looks good to me! Incremental review on 8c2b370 in 12 seconds

More details

• Looked at 13 lines of code in 1 files
• Skipped 1 files when reviewing.
• Skipped posting 1 drafted comments based on config settings.

1. pyproject.toml:42

• Draft comment:
  The version update for gptme-rag from ^0.2.1 to ^0.3.0 is consistent with the PR description and reflects the changes made in the RAG configuration. This is appropriate given the context of the PR.
• Reason this comment was not posted:
  Confidence changes required: 0%
  The version update for gptme-rag from ^0.2.1 to ^0.3.0 in pyproject.toml is consistent with the PR description. This change is appropriate given the context of the PR, which aims to simplify configuration and initialization. No issues found here.

Workflow ID: wflow_SD5rmOyP1KLHo247

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.