commands/cluster: use pipeline to execute split commands

[utkarshgupta137]

Pull Request check-list

• Does $ tox pass with this change (including linting)?
• Do the CI tests pass with this change (enable it first in your forked repo and wait for the github action build to finish)?
• Is the new or changed code fully tested?
• Is a documentation update included (if this change modifies existing APIs, or introduces new ones)?
• Is there an example added to the examples folder (if applicable)?
• Was the change added to CHANGES file?

Description of change

Since commands are split by slots, using a pipeline to send keys from the same node together provides a massive speedup, especially for the sync version.

Benchmarks using a batch of MGET 100 keys, MSET 100 keys, DEL 100 keys:

• Sync (10 batches): 6.34s -> 0.26s
• Async (100 batches): 9.34s -> 3.15s

Other changes:

• allow passing target_nodes to pipeline commands
• pass target_nodes for split commands
• move READ_COMMANDS to commands/cluster to avoid import cycle
• add types to list_or_args

[codecov-commenter]

Codecov Report

Merging #2230 (41016ff) into master (bedf3c8) will decrease coverage by 0.00%.
The diff coverage is 94.11%.

@@            Coverage Diff             @@
##           master    #2230      +/-   ##
==========================================
- Coverage   91.83%   91.82%   -0.01%     
==========================================
  Files         108      108              
  Lines       27690    27685       -5     
==========================================
- Hits        25429    25423       -6     
- Misses       2261     2262       +1     

Impacted Files	Coverage Δ
redis/cluster.py	90.08% <83.33%> (+0.08%)	⬆️
redis/asyncio/cluster.py	89.96% <85.71%> (+0.05%)	⬆️
redis/commands/__init__.py	100.00% <100.00%> (ø)
redis/commands/cluster.py	93.64% <100.00%> (-0.57%)	⬇️
redis/commands/helpers.py	87.36% <100.00%> (+0.27%)	⬆️
tests/test_cluster.py	96.79% <0.00%> (-0.12%)	⬇️
redis/asyncio/connection.py	83.85% <0.00%> (+0.11%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update bedf3c8...41016ff. Read the comment docs.

[dvora-h]

@utkarshgupta137 Thanks! These things really improve the project.