[#63] Fix HTTP transport by removing duplication of base transport

[rvoh-emccaleb]

Issue

#63

Explanation of bug

The http.HTTPTransport duplicated much of the http.baseTransport's fields and methods.

This resulted in a deadlock caused by a non-nil channel being waited on for a receive and never receiving anything because we have a nil handler. The handler is nil because the protocol is setting the message handler using http.HTTPTransport.SetMessageHandler(), which did not have the same outcome as http.HTTPTransport.baseTransport.SetMessageHandler(), due to the duplicated fields and methods in the two structs.

Solution

Removed the duplication of http.baseTransport fields and methods from the http.HTTPTransport.

I can now start up the server, locally, and observe the following:

% curl -v -X POST http://localhost:8080/mcp \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "method": "initialize",
    "id": 1,
    "params": {}
  }'
Note: Unnecessary use of -X or --request, POST is already inferred.
* Host localhost:8080 was resolved.
* IPv6: ::1
* IPv4: 127.0.0.1
*   Trying [::1]:8080...
* Connected to localhost (::1) port 8080
> POST /mcp HTTP/1.1
> Host: localhost:8080
> User-Agent: curl/8.7.1
> Accept: */*
> Content-Type: application/json
> Content-Length: 85
> 
* upload completely sent off: 85 bytes
< HTTP/1.1 200 OK
< Content-Type: application/json
< Date: Mon, 27 Jan 2025 18:44:19 GMT
< Content-Length: 217
< 
* Connection #0 to host localhost left intact
{"id":1,"jsonrpc":"2.0","result":{"capabilities":{"prompts":{"listChanged":false},"resources":{"listChanged":false},"tools":{"listChanged":false}},"protocolVersion":"2024-11-05","serverInfo":{"name":"","version":""}}}% 

Prior to the changes, this would have hung indefinitely (deadlock).

[rvoh-emccaleb]

Accidentally deleted the branch on my fork -- this PR is still open

[Chrisbattarbee]

Hey @rvoh-emccaleb ! Thanks for the contribution, currently a little strapped for time but hoping to review this tomorrow if that works for you? Same with the other PR

[rvoh-emccaleb]

@Chrisbattarbee Sounds good!

No rush on timing -- I'm currently running off of my fork. Sorry that these PRs got a little muddied with the histories (the stuff we see above) -- I mismanaged my branches, and now there is a sequential ordering to these PRs. First this one, then the other one.

[rvoh-emccaleb]

Temporarily have to close this due to company OSS policies