Update dependency @elastic/elasticsearch to v7.13.0-canary.4 (7.x)

[renovate]

This PR contains the following updates:

Package	Change	Age	Adoption	Passing	Confidence
@elastic/elasticsearch (source)	7.13.0-canary.1 -> 7.13.0-canary.4

---

Release Notes

elastic/elasticsearch-js

v7.13.0-canary.4

Compare Source

v7.13.0-canary.3

Compare Source

v7.13.0-canary.2

Compare Source

---

Configuration

📅 Schedule: At any time (no schedule defined).

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻️ Rebasing: Renovate will not automatically rebase this PR, because other commits have been found.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box.

---

This PR has been generated by WhiteSource Renovate. View repository job log here.

[elasticmachine]

Pinging @elastic/kibana-operations (Team:Operations)

[spalger]

@delvedor it looks like @elastic/elasticsearch-canary@7.13.0-canary.3 has broken types as TS is complaining that the Client class is only a type, and not a concrete class that can be instantiated, which is actually the case.

import { Client } from '@elastic/elasticsearch';

from packages/kbn-es-archiver/src/cli.ts

[delvedor]

@spalger thanks for the ping! I've published an update with the fix.

[spalger]

@elastic/kibana-core Would someone mind taking a look at the type failures here?

[pgayvallet]

@spalger I can fix the errors in this PR, now that would force to do the same thing on master once these changes lands in the 8.0.0 canary version we're using on master

Not sure why these type changes are in @elastic/elasticsearch-canary@7.13.0-canary.4 and not in @elastic/elasticsearch-canary@^8.0.0-canary.4? @delvedor?

Or are those real divergences in the ES API between 8.0 and 7.13 (kinda doubt it)?

[pgayvallet]

95f9a03 fixes the core types, however that also causes lot of other type failure to appear in plugins code now.

Should I continue on fixing? I feel like we need to do that on master and then backport, somehow.

[spalger]

Let's see if this divergence between 7.x and master is intentional, I really don't think we should diverge this much if we can avoid it.

[mshustov]

@spalger there is a PR for master #97360. I will send it as soon as I finish with #97222 (Monday, likely)

[mshustov]

closed in favor of #98266 with back port to 7.x