fix(compiler): refresh jit memory size for gas

[abmcar]

refresh memory size from the instance before expansion gas is computed normalize offset/size handling for memory-touching ops to avoid gas drift

1. Does this PR affect any open issues?(Y/N) and add issue references (e.g. "fix #123", "re #123".):

• N
• Y

2. What is the scope of this PR (e.g. component or file name):

3. Provide a description of the PR(e.g. more details, effects, motivations or doc link):

• Affects user behaviors
• Contains CI/CD configuration changes
• Contains documentation changes
• Contains experimental features
• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Other

4. Are there any breaking changes?(Y/N) and describe the breaking changes(e.g. more details, motivations or doc link):

• N
• Y

5. Are there test cases for these changes?(Y/N) select and add more details, references or doc links:

• Unit test
• Integration test
• Benchmark (add benchmark stats below)
• Manual test (add detailed scripts or steps below)
• Other

6. Release note

None

[Copilot]

Pull request overview

Updates EVM MIR compilation to reduce gas metering drift by normalizing offset/size handling for memory-touching operations and ensuring memory size used for expansion gas is refreshed from the instance.

Changes:

• Normalize offset+size pairs via normalizeOffsetWithSize(...) for multiple memory-touching op handlers (avoids incorrect trapping / gas drift when size == 0).
• Use UINT64_MAX fallback normalization for several runtime calls where non-64-bit operands should not immediately trap (e.g., CALLDATALOAD, REVERT).
• Change getMemorySize() to load from the instance (and optionally update the cached variable) to avoid stale memory size during gas computation.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

handleMStore already updates the tracked memory size via expandMemoryIR(...) when expansion happens, and the store itself cannot change EVMInstance::MemorySize. The new reloadMemorySizeFromInstance() therefore adds an extra instance load with no observable effect. Consider removing it to avoid unnecessary work on a hot path, or document why mstore uniquely needs a post-op refresh.

Suggested change

reloadMemorySizeFromInstance();

[Copilot]

getMemorySize() now always loads the size from the instance and additionally writes it into MemorySizeVar, but MemorySizeVar is not read anywhere in this translation unit anymore. This both adds an extra dassign and may regress performance vs the previous cached-variable load. Consider either (a) restoring the cached-variable fast path by keeping MemorySizeVar as the canonical source and refreshing it only when runtime calls can grow memory, or (b) removing the MemorySizeVar updates if the intent is to always read from the instance.

[Copilot]

Typo in comment: duplicated "is" in "when offset/size is is very large".

Suggested change

	// when offset/size is is very large.
	// when offset/size is very large.