Introduce Error Handling for `createChatCompletion`

[your-highness]

Recently, we are getting errors from OpenAI API which result in non gracefully exited threads. IMO it would be great to introduce some simple error handling like described in https://www.npmjs.com/package/openai?activeTab=readme#Usage

Working on this right now. Pull request will follow.

[your-highness]

It's been done in processGraphResponse() already:

chatgpt-mattermost-bot/src/process-graph-response.js

Lines 29 to 48 in 8c560a4

	try {
	const sanitized = JSON.parse(graphContent)
	const fileId = await jsonToFileId(JSON.stringify(sanitized), channelId)
	const pre = content.substring(0, replaceStart)
	const post = content.substring(replaceEnd)
	if (post.trim().length < 1){
	result.message = pre
	} else {
	result.message = `${pre} [see attached image] ${post}`
	}
	result.props = {originalMessage: content}
	result.fileId = fileId
	} catch (e) {
	log.error(e)
	log.error(`The input was:\n\n${graphContent}`)
	}
	}

But thread completion still needs some error handling. PR will cover this.