feat: StreamableHTTPServerTransport implement

[gylove1994]

Motivation and Context

To support streamable-http in mcp-2025-03-26

How Has This Been Tested?

The StreamableHTTPServerTransport has been extensively tested with the following scenarios:

1. Session Management

• Session ID generation and validation
• Session ID inclusion in response headers
• Rejection of invalid session IDs

2. Request Handling

• GET requests with and without proper Accept headers
• POST requests for JSON-RPC messages
• Handling of batch messages
• DELETE requests for session closure
• Validation of Content-Type headers

3. Message Replay

• Message replay functionality using Last-Event-ID
• Proper message sequencing and delivery

4. Message Targeting

• Correct routing of response messages to requesting connections
• Verification that responses aren't broadcast to all connections

5. Error Handling

• Invalid JSON data processing
• Invalid JSON-RPC message handling
• Proper error response formatting

All tests use mock HTTP requests and responses to simulate the StreamableHTTP specification.

Breaking Changes

None

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

Related issues: #220

[wjw99830]

Is it proper to maintain id in transport layer? It seems like a part of an application layer.

[gylove1994]

Session Management is part of the specification. See https://spec.modelcontextprotocol.io/specification/2025-03-26/basic/transports/#session-management