lsp: split monolithic server.go into modules; add configurable max tokens and context strategies (minimal|window|file-on-new-func|always-full); provide flags/env fallbacks; add unit tests for helpers and context; update README; remove obsolete files

1 files changed, 23 insertions, 17 deletions

diff --git a/IDEAS.md b/IDEAS.md
index 8bf8819..255e0db 100644
--- a/IDEAS.md
+++ b/IDEAS.md
@@ -1,35 +1,41 @@
 # Ideas
 
-## Refactor
+## Code quality
 
-* Refactor existing code in a more modular way, especially ./internal/lsp/server.go
+### Refactor
+
+* [ ] Refactor existing code in a more modular way
+* [ ] Add unit tests
 
 ## Features
 
-* LSP server to be used with the Helix text editor
-* Code completion using LLMs
-* Code generation using LLMs text
-* Be a replacement for 'github copilot cli'
+### Improvements
+
+### New features
+
+* [ ] Resolve diagnostics code action feature
+* [X] LSP server to be used with the Helix text editor
+* [X] Code completion using LLMs
+* [X] Text completion in general
+* [/] Code generation using LLMs text
+* [ ] Be a replacement for 'github copilot cli'
 
 Be able to select code blocks and perform code actions on them
 
-* Commenting exiting code
-* Code refactoring
+* [ ] Commenting exiting code
+* [ ] Code refactoring
 
 Be able to chat with the LLM
 
-* Have a dialog with the LLM, like in lsp-ai
+* [ ] Have a dialog with the LLM, like in lsp-ai
 
 Be able to switch LLMs. 
 
-* Ollama local LLM models (e.g. Qwen Coder vs Deepseek-R1 for different purposes)
-* OpenAI models
-* Claude models
-* Gemini models
-
-MSPintegration? Can't think of a specic use case right now but it may be interesting!
+* [ ] Ollama local LLM models (e.g. Qwen Coder vs Deepseek-R1 for different purposes)
+* [ ] OpenAI models
+* [ ] Claude models
+* [ ] Gemini models
 
 ## More
 
-* Use of different LLM based on usage/use case
-* Useful: https://deepwiki.com/helix-editor/helix/4.3-language-server-protocol
+* [ ] Useful: https://deepwiki.com/helix-editor/helix/4.3-language-server-protocol