Improve tool calling response handling

[MichaelAnders]

Summary

• Add fallback parsing for Ollama models that return tool calls as JSON text in message content instead of using the structured tool_calls field
• Return tool results directly to CLI instead of making a follow-up LLM call, reducing latency and preventing hallucinated rewrites of output
• Add dedicated Glob tool returning plain text (one path per line) instead of JSON, with workspace_list accepting both pattern and patterns
• Clarify why Glob is not aliased to workspace_list (format mismatch)

Files Changed

File	Change
src/clients/ollama-utils.js	+69 lines — fallback JSON parsing for tool calls in message content
src/orchestrator/index.js	+47/-1 lines — return tool results directly, skip follow-up LLM call
src/tools/index.js	+4 lines — register new Glob tool
src/tools/indexer.js	+50/-3 lines — Glob tool impl, dual-param workspace_list

Test plan

• Verify Ollama models that return tool calls as JSON text are correctly parsed
• Verify tool results are returned directly without a follow-up LLM call
• Verify Glob tool returns one file path per line (plain text)
• Verify workspace_list accepts both pattern and patterns parameters

Found some issues, removing for now