Python: added num_records parameter to text memory skill (microsoft#2236)

### Motivation and Context

<!-- Thank you for your contribution to the semantic-kernel repo!
Please help reviewers and future users, providing the following
information:
  1. Why is this change required?
  2. What problem does it solve?
  3. What scenario does it contribute to?
  4. If it fixes an open issue, please link to the issue here.
-->
The `text_memory_skill` in the core skills was setup to only return 1
record, while the underlying memories all have support for multiple, so
I added a parameter to the recall function to use that.

### Description

<!-- Describe your changes, the overall approach, the underlying design.
These notes will help understanding how your code works. Thanks! -->
Added parameter, passed that on to the memory call.
### Contribution Checklist

<!-- Before submitting this PR, please make sure: -->

- [x] The code builds clean without any errors or warnings
- [x] The PR follows the [SK Contribution
Guidelines](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md)
and the [pre-submission formatting
script](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md#development-scripts)
raises no violations
- [x] All unit tests pass, and I have added new tests where possible
- [x] I didn't break anyone 😄

---------

Co-authored-by: Shawn Callegari <[email protected]>
Co-authored-by: Abby Harrison <[email protected]>

Author: 3 people

python/semantic_kernel/core_skills/text_memory_skill.py

@@ -1,4 +1,5 @@
 # Copyright (c) Microsoft. All rights reserved.
+import json
 
 from semantic_kernel.orchestration.sk_context import SKContext
 from semantic_kernel.sk_pydantic import PydanticField
@@ -9,8 +10,10 @@ class TextMemorySkill(PydanticField):
     COLLECTION_PARAM = "collection"
     RELEVANCE_PARAM = "relevance"
     KEY_PARAM = "key"
+    LIMIT_PARAM = "limit"
     DEFAULT_COLLECTION = "generic"
     DEFAULT_RELEVANCE = 0.75
+    DEFAULT_LIMIT = 1
 
     # @staticmethod
     @sk_function(
@@ -28,6 +31,11 @@ class TextMemorySkill(PydanticField):
         description="The relevance score, from 0.0 to 1.0; 1.0 means perfect match",
         default_value=DEFAULT_RELEVANCE,
     )
+    @sk_function_context_parameter(
+        name=LIMIT_PARAM,
+        description="The maximum number of relevant memories to recall.",
+        default_value=DEFAULT_LIMIT,
+    )
     async def recall_async(self, ask: str, context: SKContext) -> str:
         """
         Recall a fact from the long term memory.
@@ -39,10 +47,11 @@ async def recall_async(self, ask: str, context: SKContext) -> str:
         Args:
             ask -- The question to ask the memory
             context -- Contains the 'collection' to search for information
-                and the 'relevance' score to use when searching
+                , the 'relevance' score to use when searching
+                and the 'limit' of relevant memories to retrieve.
 
         Returns:
-            The nearest item from the memory store
+            The nearest item from the memory store as a string or empty string if not found.
         """
         if context.variables is None:
             raise ValueError("Context has no variables")
@@ -65,15 +74,25 @@ async def recall_async(self, ask: str, context: SKContext) -> str:
         if relevance is None or str(relevance).strip() == "":
             relevance = TextMemorySkill.DEFAULT_RELEVANCE
 
+        limit = (
+            context.variables[TextMemorySkill.LIMIT_PARAM]
+            if context.variables.contains_key(TextMemorySkill.LIMIT_PARAM)
+            else TextMemorySkill.DEFAULT_LIMIT
+        )
+        if limit is None or str(limit).strip() == "":
+            limit = TextMemorySkill.DEFAULT_LIMIT
+
         results = await context.memory.search_async(
-            collection, ask, min_relevance_score=float(relevance)
+            collection=collection,
+            query=ask,
+            limit=int(limit),
+            min_relevance_score=float(relevance),
         )
         if results is None or len(results) == 0:
             if context.log is not None:
                 context.log.warning(f"Memory not found in collection: {collection}")
             return ""
-
-        return results[0].text if results[0].text is not None else ""
+        return results[0].text if limit == 1 else json.dumps([r.text for r in results])
 
     @sk_function(
         description="Save information to semantic memory",