fix: update server_version

[obycode]

The /v2/info endpoint is updated to report "subnet-node" instead of
"stacks-node".

[codecov-commenter]

Codecov Report

Merging #233 (347fc31) into master (409b9be) will decrease coverage by 1.21%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #233      +/-   ##
==========================================
- Coverage   93.47%   92.26%   -1.21%     
==========================================
  Files           6        6              
  Lines         337      336       -1     
==========================================
- Hits          315      310       -5     
- Misses         22       26       +4     

see 2 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[jbencin]

I have a few comments/questions, some of which are not directly related to the changes in this PR:

• The environment variable STACKS_NODE_VERSION still shows up in Dockerfiles in this repo. I don't think it's ever used, but these occurrences should probably be removed
• Why do we repeat the logic to generate the version string in both main.rs and rpc.rs? Can't we just call the one in main.rs from rpc.rs?
• Looking at this function:

  fn version() -> String {
    stacks::version_string(
      "subnet-node",
      option_env!("SUBNET_NODE_VERSION")
        .or(option_env!("CARGO_PKG_VERSION"))
        .unwrap_or("0.0.0.0"),
    )
  }

  I don't understand why we would ever want to return CARGO_PKG_VERSION (which is still 0.0.1 and not useful). If we don't know SUBNET_NODE_VERSION, shouldn't we give a response indicating that rather than something misleading?

[obycode]

• The environment variable STACKS_NODE_VERSION still shows up in Dockerfiles in this repo. I don't think it's ever used, but these occurrences should probably be removed

Good catch. I think these ARGs should be renamed to SUBNET_NODE_VERSION as well.

• Why do we repeat the logic to generate the version string in both main.rs and rpc.rs? Can't we just call the one in main.rs from rpc.rs?

Since main.rs is in test/stacks-node, I don't think we want to create a reference in that direction, but it could make sense to move a shared function somewhere and use it from both places.

I don't understand why we would ever want to return CARGO_PKG_VERSION (which is still 0.0.1 and not useful). If we don't know SUBNET_NODE_VERSION, shouldn't we give a response indicating that rather than something misleading?

I think this is reasonable. We may even want to make CARGO_PKG_VERSION the default once we've setup a proper release process for this repo (that's what Clarinet does for example). That version number in the Cargo.toml should eventually be incremented by our release process.

[obycode]

I'm going to leave those two calls to version_string as-is for the time being.

[jbencin]

Looks good, we can address those other things I mentioned in the future