docs: update README to document new AI providers and features

[ksylvan]

docs: update README to document new AI providers and features

Summary

This PR revises the README documentation to thoroughly detail the addition of new AI providers, features, and tools to Fabric. The changes focus on improving discoverability and usability by documenting previously undocumented capabilities.

Files Changed

README.md

Major documentation additions to cover:

• Supported AI providers list (native and OpenAI-compatible)
• Dry run mode feature
• Ollama compatibility mode for the REST API server
• Available prompt strategies with descriptions
• New generate_changelog tool documentation
• Updated table of contents with new section links

cmd/generate_changelog/incoming/1925.txt

New file containing changelog entry for this PR, following the project's changelog generation format; generated by the following command:

generate_changelog --ai-summarize --incoming-pr 1925
Commit created successfully. Please review and push manually.
Successfully created incoming changelog entry: cmd/generate_changelog/incoming/1925.txt

Code Changes

Table of Contents Updates

Added navigation links for new sections:

+    - [Supported AI Providers](#supported-ai-providers)
...
+    - [Dry Run Mode](#dry-run-mode)
...
+    - [Ollama Compatibility Mode](#ollama-compatibility-mode)
...
+      - [Available Strategies](#available-strategies)
...
+    - [`generate_changelog`](#generate_changelog)

Supported AI Providers Section

Added a comprehensive list of 25+ supported AI providers, categorized into:

• Native Integrations: OpenAI, Anthropic, Google Gemini, Ollama, Azure OpenAI, Amazon Bedrock, Vertex AI, LM Studio, Perplexity
• OpenAI-Compatible Providers: Abacus, AIML, Cerebras, DeepSeek, GitHub Models, GrokAI, Groq, Langdock, LiteLLM, MiniMax, Mistral, OpenRouter, SiliconCloud, Together, Venice AI, Z AI

Dry Run Mode Documentation

+Use `--dry-run` to preview what would be sent to the AI model without making an API call:
+
+```bash
+echo "test input" | fabric --dry-run -p summarize
+```

Ollama Compatibility Mode

Documents the ability to use Fabric as an Ollama drop-in replacement with compatible API endpoints (/api/tags, /api/chat, /api/version).

Available Strategies

Document 10 prompt strategies with descriptions:

• cot - Chain-of-Thought
• cod - Chain-of-Draft
• tot - Tree-of-Thought
• aot - Atom-of-Thought
• ltm - Least-to-Most
• self-consistent - Self-Consistency
• self-refine - Self-Refinement
• reflexion - Reflexion
• standard - Standard

generate_changelog Tool

Documents the changelog generation tool with installation instructions and feature highlights.

Reason for Changes

1. Discoverability: Many Fabric features were undocumented, making them difficult for users to discover and use
2. Onboarding: New users need clear documentation of supported AI providers during setup
3. Feature awareness: Dry run mode and Ollama compatibility are powerful features that were not documented
4. Prompt strategies: The available strategies and their usage were not explained in the README
5. Tool documentation: The generate_changelog tool lacked visibility in the main documentation

Impact of Changes

• Positive: Significantly improved documentation coverage for end users
• No breaking changes: This is purely documentation; no code behavior changes
• Improved UX: Users can now discover and leverage features they may not have known existed
• Better onboarding: New users have clearer guidance on provider options

Test Plan

1. Verify all new markdown links in table of contents resolve correctly
2. Confirm markdown renders properly on GitHub
3. Validate command examples are accurate (e.g., --dry-run, --serve, --serveOllama)
4. Check that listed AI providers match current implementation