Streaming tests added

[Adir111]

Related to this PR: kube-HPC/hkube#2009
which is related to this issue: kube-HPC/hkube#2002
Added end to end tests for new streaming scaling logic.
Warning - Do not merge until the PR is changed.

[golanha]

So little cpu might cause problems producing messages

[Adir111]

It's true, but only when we expect high message rates.
In the tests where we expect this, I’ve replaced the stateful CPU with the createAlg method:

const createAlg = async (alg, cpu) => {
        await deleteAlgorithm(alg.name, true, true)
        if (cpu) {
            alg.cpu = cpu;
        }
        await storeAlgorithms(alg);
    }

Would you prefer we increase the CPU allocation for every test's stateful configuration? As it stands, it is only modified when needed, with 0.1 being the default value.

[golanha]

Both flows have the save rate. The idea was to test different rates from each node

[Adir111]

It's a template, in the test itself the program is being initialised:

it("should satisfy the request rate of 2 statefuls, each with different rate", async () => {
await createAlg(statefull);
algList.push(statefull.name);
await createAlg(stateless);
algList.push(stateless.name);

const flow1Config = {
  programs: [
      { rate: 120, time: 50 }
  ]
};
const flow2Config = {
  flowName: "hkube_desc2",
  programs: [
      { rate: 60, time: 50 }
  ]
};
streamDifferentFlows.flowInput = combineFlows([flow1Config, flow2Config]);
...

Updated the template itself for clarity.