fix:use reentrantLock instead of spinlock

[CoffeeLatte007]

Ⅰ. Describe what this PR did

I'm sorry that I used spinLock in the wrong way before. The advantage of spinLock is that there is no need for thread context switch, but locksupport.park caused thread switch.

And SpinLock is good for locking parts to execute quickly, but we have IO operations in our changeStatus which is locking part, so to prevent CPU waste, I think ReentrantLock is more appropriate.

SpinLock can be seen as doing CAS all the time.The ReentrantLock lock the first time is CAS then it will be parked.

Ⅱ. Does this pull request fix one issue?

Ⅲ. Why don't you add test cases (unit test/integration test)?

Ⅳ. Describe how to verify it

Ⅴ. Special notes for reviews

[codecov-io]

Codecov Report

Merging #987 into develop will decrease coverage by 0.01%.
The diff coverage is 66.66%.

@@              Coverage Diff              @@
##             develop     #987      +/-   ##
=============================================
- Coverage      38.45%   38.44%   -0.02%     
  Complexity      1046     1046              
=============================================
  Files            218      218              
  Lines           8687     8681       -6     
  Branches        1081     1078       -3     
=============================================
- Hits            3341     3337       -4     
  Misses          4924     4924              
+ Partials         422      420       -2

Impacted Files	Coverage Δ	Complexity Δ
...in/java/io/seata/server/session/GlobalSession.java	80.68% <66.66%> (+0.46%)	50 <3> (ø)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 45239fe...fe0f796. Read the comment docs.

[leizhiyuan]

LGTM，

ps: if without considering more powerful lock in the future, maybe we can use ReentrantLock directly. do not need to be warpped by GlobalSessionLock

[xingfudeshi]

LGTM.

[slievrly]

LGTM