Add Faktory Enterprise Batch API support

[Otanikotani]

Summary

Adds support for Faktory Enterprise Batches, mirroring the Go client's batch API.

• New Batch class with lifecycle management (jobs(), push(), commit(), reopened())
• Client methods: batch(), batchNew(), batchOpen(), batchCommit(), batchStatus()
• Exported from the faktory module alongside existing Mutation, Job, etc.
• 10 tests using mock server infrastructure

Usage

const client = await faktory.connect();

// Create batch with callbacks
const batch = client.batch({
  description: "Import users",
  success: client.job("ImportComplete", userId),
  complete: client.job("ImportNotify", userId),
});
await batch.jobs(async () => {
  await batch.push(client.job("ImportRow", row1));
  await batch.push(client.job("ImportRow", row2));
});

// Reopen from within a batch job
const batch = await client.batchOpen(bid);
await batch.push(client.job("MoreWork", arg));
await batch.commit();

// Check status
const status = await client.batchStatus(bid);

Wire protocol

Method	Command	Response
batchNew(batch)	BATCH NEW <json>	BID string
batchOpen(bid)	BATCH OPEN <bid>	OK
batchCommit(bid)	BATCH COMMIT <bid>	OK
batchStatus(bid)	BATCH STATUS <bid>	JSON status

Test plan

• Constructor sets defaults
• toJSON() serializes callback jobs correctly
• push() throws if batch not opened
• push() throws if already committed
• commit() throws if already committed
• jobs() throws if called twice
• push() sets custom.bid on job
• jobs() sends BATCH NEW → PUSH → BATCH COMMIT in order
• batchOpen returns usable batch for reopening
• batchStatus returns parsed status object

🤖 Generated with Claude Code

[Otanikotani]

Hello @jbielick ! I want to apologize from the get go about submitting AI-assisted pull request. Please let me know if this is not something you would even consider to be added to the repository.

This PR adds .batch() functionality to node version. I tried to copy https://github.com/contribsys/faktory_worker_go for what I could. I am definitely not an expert in this, but our company uses node and we do want batching. We have something done internally already to address this, but I was thinking to share it and make it available to all.

Please let me know if something is off here!