Added ForceFlush Implementation (#272)

### Description:
Passed the force_flush function from MeteringProvider to the
AwsSpanMetricProcessor to forceFlush remaining metrics on shutdown to
the cwAgent/Collector. Compared to java/dotnet, looks like python
already flushes all the metrics and traces on shutdown even without the
force flush implementation. I'm adding it tho for consistency with other
languages.

### Tesing:
Increased the metricExporter interval and the BatchSpanProcessor delay
to 10 minutes using:
```
OTEL_METRIC_EXPORT_INTERVAL=600000 \
OTEL_BSP_SCHEDULE_DELAY=600000 \
```
With and without the force flush implementation, exiting the [sample
app](https://github.com/aws-observability/aws-otel-python-instrumentation/blob/main/sample-applications/simple-client-server/server_automatic_s3client.py)
flushed both the traces and the metrics to the collector.


By submitting this pull request, I confirm that my contribution is made
under the terms of the Apache 2.0 license.

Author: AsakerMohd

aws-opentelemetry-distro/src/amazon/opentelemetry/distro/aws_span_metrics_processor.py

@@ -1,6 +1,6 @@
 # Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
 # SPDX-License-Identifier: Apache-2.0
-from typing import Dict, Optional
+from typing import Callable, Dict, Optional
 
 from typing_extensions import override
 
@@ -45,19 +45,29 @@ class AwsSpanMetricsProcessor(SpanProcessor):
     _generator: MetricAttributeGenerator
     _resource: Resource
 
+    _force_flush_function: Callable
+
+    # no op function to act as a default function in case forceFlushFunction was
+    # not supplied to the the constructor.
+    # pylint: disable=no-self-use
+    def _no_op_function(self, timeout_millis: float = None) -> bool:
+        return True
+
     def __init__(
         self,
         error_histogram: Histogram,
         fault_histogram: Histogram,
         latency_histogram: Histogram,
         generator: MetricAttributeGenerator,
         resource: Resource,
+        force_flush_function: Callable = _no_op_function,
     ):
         self._error_histogram = error_histogram
         self._fault_histogram = fault_histogram
         self._latency_histogram = latency_histogram
         self._generator = generator
         self._resource = resource
+        self._force_flush_function = force_flush_function
 
     # pylint: disable=no-self-use
     @override
@@ -78,8 +88,8 @@ def shutdown(self) -> None:
 
     # pylint: disable=no-self-use
     @override
-    def force_flush(self, timeout_millis: int = None) -> bool:
-        return True
+    def force_flush(self, timeout_millis: float = 10_000) -> bool:
+        return self._force_flush_function(timeout_millis)
 
     def _record_metrics(self, span: ReadableSpan, attributes: BoundedAttributes) -> None:
         # Only record metrics if non-empty attributes are returned.

aws-opentelemetry-distro/src/amazon/opentelemetry/distro/aws_span_metrics_processor_builder.py

@@ -63,5 +63,10 @@ def build(self) -> AwsSpanMetricsProcessor:
         latency_histogram.name = _LATENCY
 
         return AwsSpanMetricsProcessor(
-            error_histogram, fault_histogram, latency_histogram, self._generator, self._resource
+            error_histogram,
+            fault_histogram,
+            latency_histogram,
+            self._generator,
+            self._resource,
+            self._meter_provider.force_flush,
         )