feat(shredding): add shared-shredding map placement policies

[zjw1111]

Purpose

Linked issue: #342

Add shared-shredding map column placement policies for the write path:

• Replace the previous policy-state abstraction with allocator implementations for plain, sequential, and LRU placement.
• Add per-column option fields.<column>.map.shared-shredding.column-placement-policy, defaulting to lru.
• Keep K adaptation state in MapSharedShreddingContext and let MapSharedShreddingBatchConverter read placement policy from CoreOptions when creating per-column allocators.

Tests

• cmake --build build --target paimon-common-test -j64
• cmake --build build --target paimon-core-test -j64
• env LSAN_OPTIONS=detect_leaks=0 build/debug/paimon-common-test --gtest_filter=*MapSharedShredding*
• env LSAN_OPTIONS=detect_leaks=0 build/debug/paimon-core-test --gtest_filter=CoreOptionsTest.TestDefaultValue:CoreOptionsTest.TestMapStorageLayout:*SharedShredding*
• git diff --check
• git diff --cached --check

API and Format

Adds a core option for selecting the shared-shredding map column placement policy. The physical shared-shredding map column assignment can vary by configured policy; the default policy is lru.

Documentation

No.

Generative AI tooling

Generated-by: OpenAI Codex (GPT-5)

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.