This version adds support for Azure OpenAI models. I'm not entirely happy with how each LLM provider has it's own set of params, and am investigating how to make these seem a little more unified, so this support may change in the future.
Also, Azure's content filter flags the "XXX-END-OF-SESSION-XXX" token as "sexual content", so I changed it to use "YYY" instead. I feel so protected!
The config template specified the default server version string as "SSH-2.0-OpenSSH_8.2p1 Ubuntu-4ubuntu0.3" but the SSH module automatically prepends "SSH-2.0-" to the beginning. This gave the version string returned to the client a potential fingerprint that could be used to easily identify DECEIVE honeypots. Updated the default value and added comments to document this behavior.
* 'sensor_name` is an arbitrary string that identifies the specific honeypot sensor that generated the log. Set it in the config.ini file. If not set, it will default to the honeypot system's hostname.
* 'sensor_protocol' identifies the specific protocol this honeypot sensor uses. For SSH, it's always "ssh" but as other protocols are added to DECEIVE in the future, this will have different values for their logs.
Setting a password to be "*" in the config file will cause the server to accept any password the client provides for that account, including an empty password.
The config file now contains a new "system_prompt" value in the [llm] section. This would be the same for all DECEIVE instances, and configures how the emulation itself will act. The honeypot administrator should mostly keep this intact. The prompt.txt file now focuses more on what type of system to emulate, and optional details such as valid users, contents to stage on the system, etc.
