Diagnostics are poor when a wildcard is used with beneathPath and the identified field doesn't exist everywhere in the payload

[wilkinsona]

See #714 for some background. When a field path containing a wildcard is used with beneathPath and the field doesn't exist everywhere in the payload, the resulting error message is the following:

Caused by: com.fasterxml.jackson.databind.exc.InvalidDefinitionException: No serializer found for class java.lang.Object and no properties discovered to create BeanSerializer (to avoid exception, disable SerializationFeature.FAIL_ON_EMPTY_BEANS)

An error explaining that the identified sections of the payload do not have a common structure should be thrown instead.

[wilkinsona]

I can't reproduce the problem described in #714. This test passes:

@Test
public void extractMapSubsectionWithVaryingStructureFromInconsistentJsonMap()
        throws JsonParseException, JsonMappingException, IOException {
    this.thrown.expect(PayloadHandlingException.class);
    this.thrown.expectMessage(
            "The following non-optional uncommon paths were found: [*.thread.content, *.thread.count]");
    new FieldPathPayloadSubsectionExtractor("*.thread").extractSubsection(
            "{ \"number\":\"123\", \"link\":{ \"scheme\":\"scheme\", \"url\":\"url\" }, \"chat\":{ \"message\":\"send message\", \"id\":\"id\", \"thread\":{ \"content\":\"thread content\", \"count\":1 } } }"
                    .getBytes(),
            MediaType.APPLICATION_JSON);
}

@ddaaac Can you please provide a complete and minimal sample that reproduces the error message that you reported in #714?

[ddaaac]

In my PR 71577e6, there is a test case. With some modifications to this case, following case would be not work.
And this exception seems to be thrown when the field descriptors exist.

@Test
	public void extractSubSectionWithWildcardAndFieldDescriptors() throws IOException {
		byte[] extractedPayload = new FieldPathPayloadSubsectionExtractor("*.d")
				.extractSubsection("{\"a\":{\"b\":1},\"c\":{\"d\":{\"e\":1,\"f\":2}}}".getBytes(),
								   MediaType.APPLICATION_JSON,
								   Arrays.asList(new FieldDescriptor("e").optional(), new FieldDescriptor("f").optional()));
		Map<String, Object> extracted = new ObjectMapper().readValue(extractedPayload, Map.class);
		assertThat(extracted.size()).isEqualTo(2);
		assertThat(extracted).containsOnlyKeys("e", "f");
	}