Core: Add UTs for ManifestEntriesTableTaskParser class

[pbajpai21]

Added UTs for ManifestEntriesTableTaskParser, which previously had no test coverage.

The test file TestManifestEntriesTableTaskParser mirrors the existing test
TestFilesTableTaskParser:

• nullCheck — validates the toJson/fromJson precondition guards.
• invalidJsonNode — validates rejection of non-object JSON nodes.
• testParser — full round-trip through ScanTaskParser (task → JSON → task),
  asserting both the serialized JSON and field-by-field task equality.

[pbajpai21]

@ebyhr @szehon-ho Please review the PR when you have time. Thank you.

[mkroll-db]

I'd suggest 2 minor improvements.

[mkroll-db]

Two suggestions to strengthen the round-trip test:

1. Use a projection that differs from the table schema.
2. Include two partition specs to verify collection serialization and deserialization.

Suggested change

	private BaseEntriesTable.ManifestReadTask createTask() {
	Schema schema = TestBase.SCHEMA;
	HadoopFileIO fileIO = new HadoopFileIO();
	fileIO.initialize(ImmutableMap.of("k1", "v1", "k2", "v2"));
	Map<Integer, PartitionSpec> specsById =
	PartitionUtil.indexSpecs(
	Arrays.asList(PartitionSpec.builderFor(schema).bucket("data", 16).build()));
	ManifestFile manifestFile = TestManifestFileParser.createManifestFile();
	return new BaseEntriesTable.ManifestReadTask(
	schema, fileIO, specsById, manifestFile, schema, Expressions.equal("id", 1));
	}
	private String taskJson() {
	return "{\"task-type\":\"manifest-entries-task\","
	+ "\"schema\":{\"type\":\"struct\",\"schema-id\":0,\"fields\":[{"
	+ "\"id\":3,\"name\":\"id\",\"required\":true,\"type\":\"int\"},"
	+ "{\"id\":4,\"name\":\"data\",\"required\":true,\"type\":\"string\"}]},"
	+ "\"file-io\":{\"io-impl\":\"org.apache.iceberg.hadoop.HadoopFileIO\","
	+ "\"properties\":{\"k1\":\"v1\",\"k2\":\"v2\"}},"
	+ "\"partition-specs\":[{\"spec-id\":0,\"fields\":[{"
	+ "\"name\":\"data_bucket\",\"transform\":\"bucket[16]\",\"source-id\":4,\"field-id\":1000}]}],"
	+ "\"manifest-file\":{\"path\":\"/path/input.m0.avro\","
	+ "\"length\":5878,\"partition-spec-id\":0,\"content\":0,\"sequence-number\":1,\"min-sequence-number\":2,"
	+ "\"added-snapshot-id\":12345678901234567,"
	+ "\"added-files-count\":1,\"existing-files-count\":3,\"deleted-files-count\":0,"
	+ "\"added-rows-count\":10,\"existing-rows-count\":30,\"deleted-rows-count\":0,"
	+ "\"partition-field-summary\":[{\"contains-null\":true,\"contains-nan\":false,"
	+ "\"lower-bound\":\"0A000000\",\"upper-bound\":\"64000000\"}],\"key-metadata\":\"DB030000\","
	+ "\"first-row-id\":10},"
	+ "\"projection\":{\"type\":\"struct\",\"schema-id\":0,\"fields\":[{"
	+ "\"id\":3,\"name\":\"id\",\"required\":true,\"type\":\"int\"},"
	+ "{\"id\":4,\"name\":\"data\",\"required\":true,\"type\":\"string\"}]},"
	+ "\"residual-filter\":{\"type\":\"eq\",\"term\":\"id\",\"value\":1}}";
	}
	private BaseEntriesTable.ManifestReadTask createTask() {
	Schema schema = TestBase.SCHEMA;
	Schema projection = schema.select("data");
	HadoopFileIO fileIO = new HadoopFileIO();
	fileIO.initialize(ImmutableMap.of("k1", "v1", "k2", "v2"));
	Map<Integer, PartitionSpec> specsById =
	PartitionUtil.indexSpecs(
	Arrays.asList(
	PartitionSpec.builderFor(schema).bucket("data", 16).build(),
	PartitionSpec.builderFor(schema).withSpecId(1).bucket("id", 4).build()));
	ManifestFile manifestFile = TestManifestFileParser.createManifestFile();
	return new BaseEntriesTable.ManifestReadTask(
	schema, fileIO, specsById, manifestFile, projection, Expressions.equal("id", 1));
	}
	private String taskJson() {
	return "{\"task-type\":\"manifest-entries-task\","
	+ "\"schema\":{\"type\":\"struct\",\"schema-id\":0,\"fields\":[{"
	+ "\"id\":3,\"name\":\"id\",\"required\":true,\"type\":\"int\"},"
	+ "{\"id\":4,\"name\":\"data\",\"required\":true,\"type\":\"string\"}]},"
	+ "\"file-io\":{\"io-impl\":\"org.apache.iceberg.hadoop.HadoopFileIO\","
	+ "\"properties\":{\"k1\":\"v1\",\"k2\":\"v2\"}},"
	+ "\"partition-specs\":[{\"spec-id\":0,\"fields\":[{"
	+ "\"name\":\"data_bucket\",\"transform\":\"bucket[16]\",\"source-id\":4,\"field-id\":1000}]},"
	+ "{\"spec-id\":1,\"fields\":[{"
	+ "\"name\":\"id_bucket\",\"transform\":\"bucket[4]\",\"source-id\":3,\"field-id\":1000}]}],"
	+ "\"manifest-file\":{\"path\":\"/path/input.m0.avro\","
	+ "\"length\":5878,\"partition-spec-id\":0,\"content\":0,\"sequence-number\":1,\"min-sequence-number\":2,"
	+ "\"added-snapshot-id\":12345678901234567,"
	+ "\"added-files-count\":1,\"existing-files-count\":3,\"deleted-files-count\":0,"
	+ "\"added-rows-count\":10,\"existing-rows-count\":30,\"deleted-rows-count\":0,"
	+ "\"partition-field-summary\":[{\"contains-null\":true,\"contains-nan\":false,"
	+ "\"lower-bound\":\"0A000000\",\"upper-bound\":\"64000000\"}],\"key-metadata\":\"DB030000\","
	+ "\"first-row-id\":10},"
	+ "\"projection\":{\"type\":\"struct\",\"schema-id\":0,\"fields\":[{"
	+ "\"id\":4,\"name\":\"data\",\"required\":true,\"type\":\"string\"}]},"
	+ "\"residual-filter\":{\"type\":\"eq\",\"term\":\"id\",\"value\":1}}";
	}

[pbajpai21]

@mkroll-db, Thank you, great suggestions - applied both. Using a projection that differs from the table schema (data only) and adding a second partition spec makes the round-trip actually verify that the projection and the spec collection are serialized/deserialized independently, rather than passing coincidentally.

Verified locally: test green, spotless and checkstyle clean. Appreciate the review...

[mkroll-db]

LGTM. No more suggestions.