feat: enable zstd compression and encodings in merge tree data part

[v0y4g3r]

I hereby agree to the terms of the GreptimeDB CLA

What's changed and what's your intention?

This PR enables ZSTD compression and column-wise encodings in DataPart in merge tree memtable to save memory up to 90%.

Total memory consumption for freezing a DataBuffer with 10M rows to DataPart and iterating all batches from that DataPart:

The results

• Elapsed time means the total time for freezingDataBuffer with 10 million rows to DataPart and iterating that DataPart.

Type	Encoded part size	Elapsed time
Raw	315MB (1260%)	2.98s (102%)
Encodings only	166MB (664%)	2.91s (100%)
Zstd only	49MB(196%)	3.80s (131%)
Zstd&encodings	25MB(100%)	3.30s (113%)

Checklist

• I have written the necessary rustdoc comments.
• I have added the necessary unit tests and integration tests.
• This PR does not require documentation updates.

Refer to a related PR or issue link (optional)

• Tracking issue for the memtable improvement #2804

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 85.24%. Comparing base (d8dc93f) to head (b060629).

❗ Current head b060629 differs from pull request most recent head 69f8185. Consider uploading reports for the commit 69f8185 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3380      +/-   ##
==========================================
- Coverage   85.67%   85.24%   -0.43%     
==========================================
  Files         893      893              
  Lines      147382   147251     -131     
==========================================
- Hits       126270   125531     -739     
- Misses      21112    21720     +608     

[fengjiachun]

LGTM